hi everybody thank you for joining us today for the virtual Vertica BBC 2020 today's breakout session is entitled Vertica and en mode at the trade desk my name is su LeClair director of marketing at Vertica and I'll be your host for this webinar joining me is Ron Cormier senior Vertica database engineer at the trade desk before we begin I encourage you to submit questions or comments during the virtual session you don't have to wait just type your question or comment in the question box below the slides and click submit there will be a Q&A session at the end of the presentation we'll answer as many questions as we're able to during that time any questions that we don't address we'll do our best to answer them offline alternatively you can visit vertical forums to post your questions there after the session our engineering team is planning to join the forums to keep the conversation going also a quick reminder that you can maximize your screen by clicking the double arrow button in the lower right corner of the slide and yes this virtual session is being recorded and will be available to view on demand this week we'll send you a notification as soon as it's ready so let's get started over to you run thanks - before I get started I'll just mention that my slide template was created before social distancing was a thing so hopefully some of the images will harken us back to a time when we could actually all be in the same room but with that I want to get started uh the date before I get started in thinking about the technology I just wanted to cover my background real quick because I think it's peach to where we're coming from with vertically on at the trade desk and I'll start out just by pointing out that prior to my time in the trade desk I was a tech consultant at HP HP America and so I traveled the world working with Vertica customers helping them configure install tune set up their verdict and databases and get them working properly so I've seen the biggest and the smallest implementations and everything in between and and so now I'm actually principal database engineer straight desk and and the reason I mentioned this is to let you know that I'm a practitioner I'm working with with the product every day or most days this is a marketing material so hopefully the the technical details in this presentation are are helpful I work with Vertica of course and that is most relative or relevant to our ETL and reporting stack and so what we're doing is we're taking about the data in the Vertica and running reports for our customers and we're an ad tech so I did want to just briefly describe what what that means and how it affects our implementation so I'm not going to cover the all the details of this slide but basically I want to point out that the trade desk is a DSP it's a demand-side provider and so we place ads on behalf of our customers or agencies and ad agencies and their customers that are advertised as brands themselves and the ads get placed on to websites and mobile applications and anywhere anywhere digital advertising happens so publishers are what we think ocean like we see here espn.com msn.com and so on and so every time a user goes to one of these sites or one of these digital places and an auction takes place and what people are bidding on is the privilege of showing and add one or more ads to users and so this is this is really important because it helps fund the internet ads can be annoying sometimes but they actually help help are incredibly helpful in how we get much much of our content and this is happening in real time at very high volumes so on the open Internet there is anywhere from seven to thirteen million auctions happening every second of those seven to thirteen million auctions happening every second the trade desk bids on hundreds of thousands per second um so that gives it and anytime we did we have an event that ends up in Vertica that's that's one of the main drivers of our data volume and certainly other events make their way into Vertica as well but that wanted to give you a sense of the scale of the data and sort of how it's impacting or how it is impacted by sort of real real people in the world so um the uh let's let's take a little bit more into the workload and and we have the three B's in spades late like many many people listening to a massive volume velocity and variety in terms of the data sizes I've got some information here some stats on on the raw data sizes that we deal with on a daily basis per day so we ingest 85 terabytes of raw data per day and then once we get it into Vertica we do some transformations we do matching which is like joins basically and we do some aggregation group buys to reduce the data and make it clean it up make it so it's more efficient to consume buy our reporting layer so that matching in aggregation produces about ten new terabytes of raw data per day it all comes from the it all comes from the data that was ingested but it's new data and so that's so it is reduced quite a bit but it's still pretty pretty high high volume and so we have this aggregated data that we then run reports on on behalf of our customers so we have about 40,000 reports per day oh that's probably that's actually a little bit old and older number it's probably closer to 50 or 55,000 reports per day at this point so it's I think probably a pretty common use case for for Vertica customers it's maybe a little different in the sense that most of the reports themselves are >> reports so they're not it's not a user sitting at a keyboard waiting for the result basically we have we we have a workflow where we do the ingest we do this transform and then and then once once all the data is available for a day we run reports on behalf of our customer to let me have our customers on that that daily data and then we send the reports out you via email or we drop them in a shared location and then they they look at the reports at some later point of time so it's up until yawn we did all this work on on enterprise Vertica at our peak we had four production enterprise clusters each which held two petabytes of raw data and I'll give you some details on on how those enterprise clusters were configured in the hardware but before I do that I want to talk about the reporting workload specifically so the the reporting workload is particularly lumpy and what I mean by that is there's a bunch of work that becomes available bunch of queries that we need to run in a short period of time after after the days just an aggregation is completed and then the clusters are relatively quiet for the remaining portion of the day that's not to say they are they're not doing anything as far as read workload but they certainly are but it's much less reactivity after that big spike so what I'm showing here is our reporting queue and the spike is is when all those reports become a bit sort of ailable to be processed we can't we can't process we can't run the report until we've done the full ingest and matching and aggregation for the day and so right around 1:00 or 2:00 a.m. UTC time every day that's when we get this spike and the spike we affectionately called the UTC hump but basically it's a huge number of queries that need to be processed sort of as soon as possible and we have service levels that dictate what as soon as possible means but I think the spike illustrates our use case pretty pretty accurately and um it really as we'll see it's really well suited for pervert icky on and we'll see what that means so we've got our we had our enterprise clusters that I mentioned earlier and just to give you some details on what they look like there they were independent and mirrored and so what that means is all four clusters held the same data and we did this intentionally because we wanted to be able to run our report anywhere we so so we've got this big queue over port is big a number of reports that need to be run and we've got these we started we started with one cluster and then we got we found that it couldn't keep up so we added a second and we found the number of reports went up that we needed to run that short period of time and and so on so we eventually ended up with four Enterprise clusters basically with this with the and we'd say they were mirrored they all had the same data they weren't however synchronized they were independent and so basically we would run the the tailpipe line so to speak we would run ingest and the matching and the aggregation on all the clusters in parallel so they it wasn't as if each cluster proceeded to the next step in sync with which dump the other clusters they were run independently so it was sort of like each each cluster would eventually get get consistent and so this this worked pretty well for for us but it created some imbalances and there was some cost concerns that will dig into but just to tell you about each of these each of these clusters they each had 50 nodes they had 72 logical CPU cores a half half a terabyte of RAM a bunch of raid rated disk drives and 2 petabytes of raw data as I stated before so pretty big beefy nodes that are physical physical nodes that we held we had in our data centers we actually reached these nodes so so it was on our data center providers data centers and the these were these these were what we built our business on basically but there was a number of challenges that we ran into as we as we continue to build our business and add data and add workload and and the first one is is some in ceremony can relate to his capacity planning so we had to prove think about the future and try to predict the amount of work that was going to need to be done and how much hardware we were going to need to satisfy that work to meet that demand and that's that's just generally a hard thing to do it's very difficult to verdict the future as we can probably all attest to and how much the world has changed and even in the last month so it's a it's a very difficult thing to do to look six twelve eighteen eighteen months into the future and sort of get it right and and and what people what we tended to do is we reach or we tried to our art plans our estimates were very conservative so we overbought in a lot of cases and not only that we had to plan for the peak so we're planning for that that that point in time that those number of hours in the early morning when we had to we had all those reports to run and so that so so we ended up buying a lot of hardware and we actually sort of overbought at times and then and then as the hardware were days it would kind of come into it would come into maturity and we have our our our workload would sort of come approach matching the demand so that was one of the big challenges the next challenge is that we were running on disk you can we wanted to add data in sort of two dimensions the only dimensions that everybody can think about we wanted to add more columns to our big aggregates and we wanted to keep our big aggregates for for longer periods of time so both horizontally and vertically we wanted to expand the datasets but we basically were running out of disk there was no more disk in and it's hard to add a disc to Vertica in enterprise mode not not impossible but certainly hard and and one cannot add discs without adding compute because enterprise mode the disk is all local to each of the nodes for most most people you can do not exchange with sands and other external rays but that's there are a number of other challenges with that so um adding in order to add disk we had to add compute and that basically meant kept us out of balance we're adding more compute than we needed for the amount of disk so that was the problem certainly physical nodes getting them the order delivered racked cables even before we even start such Vertica there's lead times there and and so it's also long commitment since we like I mentioned me Lisa hardware so we were committing to these nodes these physical servers for two or three years at a time and I mentioned that can be a hard thing to do but we wanted to least to keep our capex down so we wanted to keep our aggregates for a long period of time we could have done crazy things or more exotic things to to help us with this if we had to in enterprise mode we could have started to like daisy chain clusters together and that would have been sort of a non-trivial engineering effort because we would need to then figure out how to migrate data source first to recharge the data across all the clusters and we had to migrate data from one cluster to another cluster hesitation and we would have to think about how to aggregate run queries across clusters so if you assured data set spans two clusters it would have had to sort of aggregated within each cluster maybe and then build something on top the aggregated the data from each of those clusters so not impossible things but certainly not easy things and luckily for us we started talking about two Vertica about separation of compute and storage and I know other customers were talking to Vertica as we were people had had these problems and so Vertica inyeon mode came to the rescue and what I want to do is just talk about nyan mode really briefly for for those in the audience who aren't familiar but it's basically Vertigo's answered to the separation of computing storage it allows one to scale compute and or storage separately and and this there's a number of advantages to doing that whereas in the old enterprise days when you add a compute you added stores and vice-versa now we can now we can add one or the other or both according to how we want to and so really briefly how this works this slide this figure was taken directly from the verdict and documentation and so just just to talk really briefly about how it works the taking advantage of the cloud and so in this case Amazon Web Services the elasticity in the cloud and basically we've got you seen two instances so elastic cloud compute servers that access data that's in an s3 bucket and so three three ec2 nodes and in a bucket or the the blue objects in this diagram and the difference is a couple of a couple of big differences one the data no longer the persistent storage of the data the data where the data lives is no longer on each of the notes the persistent stores of the data is in s3 bucket and so what that does is it basically solves one of our first big problems which is we were running out of disk the s3 has for all intensive purposes infinite storage so we can keep much more data there and that mostly solved one of our big problems so the persistent data lives on s3 now what happens is when a query runs it runs on one of the three nodes that you see here and assuming we'll talk about depo in a second but what happens in a brand new cluster where it's just just spun up the hardware is the query will will run on those ec2 nodes but there will be no data so those nodes will reach out to s3 and run the query on remote storage so that so the query that the nodes are literally reaching out to the communal storage for the data and processing it entirely without using any data on on the nodes themselves and so that that that works pretty well it's not as fast as if the data was local to the nodes but um what Vertica did is they built a caching layer on on each of the node and that's what the depot represents so the depot is some amount of disk that is relatively local to the ec2 node and so when the query runs on remote stores on the on the s3 data it then queues up the data for download to the nodes and so the data will get will reside in the Depot so that the next query or the subsequent subsequent queries can run on local storage instead of remote stores and that speeds things up quite a bit so that that's that's what the role of the Depot is the depot is basically a caching layer and we'll talk about the details of how we can see your in our Depot the other thing that I want to point out is that since this is the cloud another problem that helps us solve is the concurrency problem so you can imagine that these three nodes are one sort of cluster and what we can do is we can spit up another three nodes and have it point to the same s3 communal storage bucket so now we've got six nodes pointing to the same data but we've you isolated each of the three nodes so that they act as if they are their own cluster and so vertical calls them sub-clusters so we've got two sub clusters each of which has three nodes and what this has essentially done it is it doubled the concurrency doubled the number of queries that can run at any given time because we've now got this new place which new this new chunk of compute which which can answer queries and so that has given us the ability to add concurrency much faster and I'll point out that for since it's cloud and and there are on-demand pricing models we can have significant savings because when a sub cluster is not needed we can stop it and we pay almost nothing for it so that's that's really really important really helpful especially for our workload which I pointed out before was so lumpy so those hours of the day when it's relatively quiet I can go and stop a bunch of sub clusters and and I will pay for them so that that yields nice cost savings let's be on in a nutshell obviously engineers and the documentation can use a lot more information and I'm happy to field questions later on as well but I want to talk about how how we implemented beyond at the trade desk and so I'll start on the left hand side at the top the the what we're representing here is some clusters so there's some cluster 0 r e t l sub cluster and it is a our primary sub cluster so when you get into the world of eon there's primary Club questions and secondary sub classes and it has to do with quorum so primary sub clusters are the sub clusters that we always expect to be up and running and they they contribute to quorum they decide whether there's enough instances number a number of enough nodes to have the database start up and so these this is where we run our ETL workload which is the ingest the match in the aggregate part of the work that I talked about earlier so these nodes are always up and running because our ETL pipeline is always on we're internet ad tech company like I mentioned and so we're constantly getting costly running ad and there's always data flowing into the system and the matching is happening in the aggregation so that part happens 24/7 and we wanted so that those nodes will always be up and running and we need this we need that those process needs to be super efficient and so what that is reflected in our instance type so each of our sub clusters is sixty four nodes we'll talk about how we came at that number but the infant type for the ETL sub cluster the primary subclusters is I 3x large so that is one of the instance types that has quite a bit of nvme stores attached and we'll talk about that but on 32 cores 240 four gigs of ram on each node and and that what that allows us to do I should have put the amount of nvme but I think it's seven terabytes for anything me storage what that allows us to do is to basically ensure that our ETL everything that this sub cluster does is always in Depot and so that that makes sure that it's always fast now when we get to the secondary subclusters these are as mentioned secondary so they can stop and start and it won't affect the cluster going up or down so they're they're sort of independent and we've got four what we call Rhian subclusters and and they're not read by definition or technically they're not read only any any sub cluster can ingest and create your data within the database and that'll all get that'll all get pushed to the s3 bucket but logically for us they're read only like these we just most of these the work that they happen to do is read only which it is which is nice because if it's read only it doesn't need to worry about commits and we let we let the primary subclusters or ETL so close to worry about committing data and we don't have to we don't have to have the all nodes in the database participating in transaction commits so we've got a for read subclusters and we've got one EP also cluster so a total of five sub clusters each so plus they're running sixty-four nodes so that gives us a 320 node database all things counted and not all those nodes are up at the same time as I mentioned but often often for big chunks of the days most of the read nodes are down but they do all spin up during our during our busy time so for the reading so clusters we've got I three for Excel so again the I three incidents family type which has nvme stores these notes have I think three and a half terabytes of nvme per node we just rate it to nvme drives we raid zero them together and 16 cores 122 gigs of ram so these are smaller you'll notice but it works out well for us because the the read workload is is typically dealing with much smaller data sets than then the ingest or the aggregation workbook so we can we can run these workloads on on smaller instances and leave a little bit of money and get more granularity with how many sub clusters are stopped and started at any given time the nvme doesn't persist the data on it isn't persisted remember you stop and start this is an important detail but it's okay because the depot does a pretty good job in that in that algorithm where it pulls data in that's recently used and the that gets pushed out a victim is the data that's least reasons use so it was used a long time ago so it's probably not going to be used to get so we've got um five sub-clusters and we have actually got to two of those so we've got a 320 node cluster in u.s. East and a 320 node cluster in u.s. West so we've got a high availability region diversity so and their peers like I talked about before they're they're independent but but yours they are each run 128 shards and and so with that what that which shards are is basically the it's similar to segmentation when you take those dataset you divide it into chunks and though and each sub cluster can concede want the data set in its entirety and so each sub cluster is dealing with 128 shards it shows 128 because it'll give us even distribution of the data on 64 node subclusters 60 120 might evenly by 64 and so there's so there's no data skew and and we chose 128 because the sort of ginger proof in case we wanted to double the size of any of the questions we can double the number of notes and we still have no excuse the data would be distributed evenly the disk what we've done is so we've got a couple of raid arrays we've got an EBS based array that they're catalog uses so the catalog storage location and I think we take for for EBS volumes and raid 0 them together and come up with 128 gigabyte Drive and we wanted an EPS for the catalog because it we can stop and start nodes and that data will persist it will come back when the node comes up so we don't have to run a bunch of configuration when the node starts up basically the node starts it automatically joins the cluster and and very strongly there after it starts processing work let's catalog and EBS now the nvme is another raid zero as I mess with this data and is ephemeral so let me stop and start it goes away but basically we take 512 gigabytes of the nvme and we give it to the data temp storage location and then we take whatever is remaining and give it to the depot and since the ETL and the reading clusters are different instance types they the depot is is side differently but otherwise it's the same across small clusters also it all adds up what what we have is now we we stopped the purging data for some of our big a grits we added bunch more columns and what basically we at this point we have 8 petabytes of raw data in each Jian cluster and it is obviously about 4 times what we can hold in our enterprise classes and we can continue to add to this maybe we need to add compute maybe we don't but the the amount of data that can can be held there against can obviously grow much more we've also built in auto scaling tool or service that basically monitors the queue that I showed you earlier monitors for those spikes I want to see as low spikes it then goes and starts up instances one sub-collector any of the sub clusters so that's that's how that's how we we have compute match the capacity match that's the demand also point out that we actually have one sub cluster is a specialized nodes it doesn't actually it's not strictly a customer reports sub clusters so we had this this tool called planner which basically optimizes ad campaigns for for our customers and we built it it runs on Vertica uses data and Vertica runs vertical queries and it was it was wildly successful um so we wanted to have some dedicated compute and beyond witty on it made it really easy to basically spin up one of these sub clusters or new sub cluster and say here you go planner team do what you want you can you can completely maximize the resources on these nodes and it won't affect any of the other operations that were doing the ingest the matching the aggregation or the reports up so it gave us a great deal of flexibility and agility which is super helpful so the question is has it been worth it and without a doubt the answer is yes we're doing things that we never could have done before sort of with reasonable cost we have lots more data specialized nodes and more agility but how do you quantify that because I don't want to try to quantify it for you guys but it's difficult because each eon we still have some enterprise nodes by the way cost as you have two of them but we also have these Eon clusters and so they're there they're running different workloads the aggregation is different the ingest is running more on eon does the number of nodes is different the hardware is different so there are significant differences between enterprise and and beyond and when we combine them together to do the entire workload but eon is definitely doing the majority of the workload it has most of the data it has data that goes is much older so it handles the the heavy heavy lifting now the query performance is more anecdotal still but basically when the data is in the Depot the query performance is very similar to enterprise quite close when the data is not in Depot and it needs to run our remote storage the the query performance is is is not as good it can be multiples it's not an order not orders of magnitude worse but certainly multiple the amount of time that it takes to run on enterprise but the good news is after the data downloads those young clusters quickly catch up as the cache populates there of cost I'd love to be able to tell you that we're running to X the number of reports or things are finishing 8x faster but it's not that simple as you Iran is that you it is me I seem to have gotten to thank you you hear me okay I can hear you now yeah we're still recording but that's fine we can edit this so if I'm just talking to the person the support person he will extend our recording time so if you want to maybe pick back up from the beginning of the slide and then we'll just edit out this this quiet period that we have sir okay great I'm going to go back on mute and why don't you just go back to the previous slide and then come into this one again and I'll make sure that I tell the person who yep perfect and then we'll continue from there is that okay yeah sound good all right all right I'm going back on yet so the question is has it been worth it and for us the answer has been a resounding yes we're doing things that we never could have done at reasonable cost before and we got more data we've got this Y note this law has nodes and in work we're much more agile so how to quantify that um well it's not quite as simple and straightforward as you might hope I mean we still have enterprise clusters we've got to update the the four that we had at peak so we've still got two of those around and we got our two yawn clusters but they're running different workloads and they're comprised of entirely different hardware the dependence has I've covered the number of nodes is different for sub-clusters so 64 versus 50 is going to have different performance the the workload itself the aggregation is aggregating more columns on yon because that's where we have disk available the queries themselves are different they're running more more queries on more intensive data intensive queries on yon because that's where the data is available so in a sense it is Jian is doing the heavy lifting for the cluster for our workload in terms of query performance still a little anecdotal but like when the queries that run on the enterprise cluster the performance matches that of the enterprise cluster quite closely when the data is in the Depot when the data is not in a Depot and Vertica has to go out to the f32 to get the data performance degrades as you might expect it can but it depends on the curious all things like counts counts are is really fast but if you need lots of the data from the material others to realize lots of columns that can run slower I'm not orders of magnitude slower but certainly multiple of the amount of time in terms of costs anecdotal will give a little bit more quantifying here so what I try to do is I try to figure out multiply it out if I wanted to run the entire workload on enterprise and I wanted to run the entire workload on e on with all the data we have today all the queries everything and to try to get it to the Apple tab so for enterprise the the and estimate that we do need approximately 18,000 cores CPU cores all together and that's a big number but that's doesn't even cover all the non-trivial engineering work that would need to be required that I kind of referenced earlier things like starting the data among multiple clusters migrating the data from one culture to another the daisy chain type stuff so that's that's the data point now for eon is to run the entire workload estimate we need about twenty thousand four hundred and eighty CPU cores so more CPU cores uh then then enterprise however about half of those and partly ten thousand of both CPU cores would only run for about six hours per day and so with the on demand and elasticity of the cloud that that is a huge advantage and so we are definitely moving as fast as we can to being on all Aeon we have we have time left on our contract with the enterprise clusters or not we're not able to get rid of them quite yet but Eon is certainly the way of the future for us I also want to point out that uh I mean yawn is we found to be the most efficient MPP database on the market and what that refers to is for a given dollar of spend of cost we get the most from that zone we get the most out of Vertica for that dollar compared to other cloud and MPP database platforms so our business is really happy with what we've been able to deliver with Yan Yan has also given us the ability to begin a new use case which is probably this case is probably pretty familiar to folks on the call where it's UI based so we'll have a website that our customers can log into and on that website they'll be able to run reports on queries through the website and have that run directly on a separate row to get beyond cluster and so much more latent latency sensitive and concurrency sensitive so the workflow that I've described up until this point has been pretty steady throughout the day and then we get our spike and then and then it goes back to normal for the rest of the day this workload it will be potentially more variable we don't know exactly when our engineers are going to deliver some huge feature that is going to make a 1-1 make a lot of people want to log into the website and check how their campaigns are doing so we but Yohn really helps us with this because we can add a capacity so easily we cannot compute and we can add so we can scale that up and down as needed and it allows us to match the concurrency so beyond the concurrency is much more variable we don't need a big long lead time so we're really excited about about this so last slide here I just want to leave you with some things to think about if you're about to embark or getting started on your journey with vertically on one of the things that you'll have to think about is the no account in the shard count so they're kind of tightly coupled the node count we determined by figuring like spinning up some instances in a single sub cluster and getting performance smaller to finding an acceptable performance considering current workload future workload for the queries that we had when we started and so we went with 64 we wanted to you want to certainly want to increase over 50 but we didn't want to have them be too big because of course it costs money and so what you like to do things in power to so 64 nodes and then the shard count for the shards again is like the data segmentation is a new type of segmentation on the data and the start out we went with 128 it began the reason is so that we could have no skew but you know could process the same same amount of data and we wanted to future-proof it so that's probably it's probably a nice general recommendation doubleness account for the nodes the instance type and and how much people space those are certainly things you're going to consider like I was talking about we went for they I three for Excel I 3/8 Excel because they offer good good Depot stores which gives us a really consistent good performance and it is all in Depot the pretty good mud presentation and some information on on I think we're going to use our r5 or the are for instance types for for our UI cluster so much less the data smaller so much less enter this on Depot so we don't need on that nvm you stores the reader we're going to want to have a reserved a mix of reserved and on-demand instances if you're if you're 24/7 shop like we are like so our ETL subclusters those are reserved instances because we know we're going to run those 24 hours a day 365 days a year so there's no advantage of having them be on-demand on demand cost more than reserve so we get cost savings on on figuring out what we're going to run and have keep running and it's the read subclusters that are for the most part on on demand we have one of our each sub Buster's is actually on 24/7 because we keep it up for ad-hoc queries your analyst queries that we don't know when exactly they're going to hit and they want to be able to continue working whenever they want to in terms of the initial data load the initial data ingest what we had to do and now how it works till today is you've got to basically load all your data from scratch there isn't a great tooling just yet for data populate or moving from enterprise to Aeon so what we did is we exported all the data in our enterprise cluster into park' files and put those out on s3 and then we ingested them into into our first Eon cluster so it's kind of a pain we script it out a bunch of stuff obviously but they worked and the good news is that once you do that like the second yon cluster is just a bucket copy in it and so there's tools missions that can help help with that you're going to want to manage your fetches and addiction so this is the data that's in the cache is what I'm referring to here the data that's in the default and so like I talked about we have our ETL cluster which has the most recent data that's just an injected and the most difficult data that's been aggregated so this really recent data so we wouldn't want anybody logging into that ETL cluster and running queries on big aggregates to go back one three years because that would invalidate the cache the depot would start pulling in that historical data and it was our assessing that historical data and evicting the recent data which would slow things out flow down that ETL pipelines so we didn't want that so we need to make sure that users whether their service accounts or human users are connecting to the right phone cluster and I mean we just created the adventure users with IPS and target groups to palm those pretty-pretty it was definitely something to think about lastly if you're like us and you're going to want to stop and start nodes you're going to have to have a service that does that for you we're where we built this very simple tool that basically monitors the queue and stops and starts subclusters accordingly we're hoping that that we can work with Vertica to have it be a little bit more driven by the cloud configuration itself so for us all amazon and we love it if we could have it have a scale with the with the with the eight of us can take through points do things to watch out for when when you're working with Eon is the first is system table queries on storage layer or metadata and the thing to be careful of is that the storage layer metadata is replicated it's caught as a copy for each of the sub clusters that are out there so we have the ETL sub cluster and our resources so for each of the five sub clusters there is a copy of all the data in storage containers system table all the data and partitions system table so when you want to use this new system tables for analyzing how much data you have or any other analysis make sure that you filter your query with a node name and so for us the node name is less than or equal to 64 because each of our sub clusters at 64 so we limit we limit the nodes to the to the 64 et 64 node ETL collector otherwise if we didn't have this filter we would get 5x the values for counts and some sort of stuff and lastly there is a problem that we're kind of working on and thinking about is a DC table data for sub clusters that are our stops when when the instances stopped literally the operating system is down and there's no way to access it so it takes the DC table DC table data with it and so I cannot after after my so close to scale up in the morning and then they scale down I can't run DC table queries on how what performed well and where and that sort of stuff because it's local to those nodes so we're working on something so something to be aware of and we're working on a solution or an implementation to try to suck that data out of all the notes you can those read only knows that stop and start all the time and bring it in to some other kind of repository perhaps another vertical cluster so that we can run analysis and monitoring even you want those those are down that's it um thanks for taking the time to look into my presentation really do it thank you Ron that was a tremendous amount of information thank you for sharing that with everyone um we have some questions come in that I would like to present to you Ron if you have a couple min it your first let's jump right in the first one a loading 85 terabytes per day of data is pretty significant amount what format does that data come in and what does that load process look like yeah a great question so the format is a tab separated files that are Jesus compressed and the reason for that could basically historical we don't have much tabs in our data and this is how how the data gets compressed and moved off of our our bidders the things that generate most of this data so it's a PSD gzip compressed and how you kind of we kind of have how we load it I would say we have actually kind of a Cadillac loader in a couple of different perspectives one is um we've got this autist raishin layer that's homegrown managing the logs is the data that gets loaded into Vertica and so we accumulate data and then we take we take some some files and we push them to redistribute them along the ETL nodes in the cluster and so we're literally pushing the file to through the nodes and we then run a copy statement to to ingest data in the database and then we remove the file from from the nodes themselves and so it's a little bit extra data movement which you may think about changing in the future assisting we move more and more to be on well the really nice thing about this especially for for the enterprise clusters is that the copy' statements are really fast and so we the coffee statements use memory but let's pick any other query but the performance of the cautery statement is really sensitive to the amount of available memory and so since the data is local to the nodes literally in the data directory that I referenced earlier it can access that data from the nvme stores and the kabhi statement runs very fast and then that memory is available to do something else and so we pay a little bit of cost in terms of latency and in terms of downloading the data to the nose we might as we move more and more PC on we might start ingesting it directly from s3 not copying the nodes first we'll see about that what's there that's how that's how we read the data interesting works great thanks Ron um another question what was the biggest challenge you found when migrating from on-prem to AWS uh yeah so um a couple of things that come to mind the first was the baculum the data load it was kind of a pain I mean like I referenced in that last slide only because I mean we didn't have tools built to do this so I mean we had to script some stuff out and it wasn't overly complex but yes it's just a lot of data to move I mean even with starting with with two petabytes so making sure that there there is no missed data no gaps making and moving it from the enterprise cluster so what we did is we exported it to the local disk on the enterprise buses and we then we push this history and then we ingested it in ze on again Allspark X oh so it's a lot of days to move around and I mean we have to you have to take an outage at some point stop loading data while we do that final kiss-up phase and so that was that was a challenge a sort of a one-time challenge the other saying that I mean we've been dealing with a week not that we're dealing with but with his challenge was is I mean it's relatively you can still throw totally new product for vertical and so we are big advantages of beyond is allow us to stop and start nodes and recently Vertica has gotten quite good at stopping in part starting nodes for a while there it was it was it took a really long time to start to Noah back up and it could be invasive but we worked with with the engineering team with Yan Zi and others to really really reduce that and now it's not really an issue that we think that we think too much about hey thanks towards the end of the presentation you had said that you've got 128 shards but you have your some clusters are usually around 64 nodes and you had talked about a ratio of two to one why is that and if you were to do it again would you use 128 shards ah good question so that is a reference the reason why is because we wanted to future professionals so basically we wanted to make sure that the number of stars was evenly divisible by the number of nodes and you could I could have done that was 64 I could have done that with 128 or any other multiple entities for but we went with 128 is to try to protect ourselves in the future so that if we wanted to double the number of nodes in the ECL phone cluster specifically we could have done that so that was double from 64 to 128 and then each node would have happened just one chart that it had would have to deal with so so no skew um the second part of question if I had to do it if I had to do it over again I think I would have done I think I would have stuck with 128 we still have I mean so we either running this cluster for more than 18 months now I think especially in USC and we haven't needed to increase the number of nodes so in that sense like it's been a little bit extra overhead having more shards but it gives us the peace of mind that we can easily double that and not have to worry about it so I think I think everyone is a nice place to start and you may even consider a three to one or four to one if if you're if you're expecting really rapid growth that you were just getting started with you on and your business and your gates that's a small now but what you expect to have them grow up significantly less powerful green thank you Ron that's with all the questions that we have out there for today if you do have others please feel free to send them in and we will get back to you and we'll respond directly via email and again our engineers will be available on the vertical forums where you can continue the discussion with them there I want to thank Ron for the great presentation and also the audience for your participation in questions please note that a replay of today's event and a copy of the slides will be available on demand shortly and of course we invite you to share this information with your colleagues as well again thank you and this concludes this webinar and have a great day you

from the cube studios in palo alto in boston bringing you data-driven insights from the cube and etr this is breaking analysis with dave vellante nvidia wants to completely transform enterprise computing by making data centers run 10x faster at one tenth the cost and video's ceo jensen wang is crafting a strategy to re-architect today's on-prem data centers public clouds and edge computing installations with a vision that leverages the company's strong position in ai architectures the keys to this end-to-end strategy include a clarity of vision massive chip design skills a new arm-based architecture approach that integrates memory processors i o and networking and a compelling software consumption model even if nvidia is unsuccessful at acquiring arm we believe it will still be able to execute on this strategy by actively participating in the arm ecosystem however if its attempts to acquire arm are successful we believe it will transform nvidia from the world's most valuable chip company into the world's most valuable supplier of integrated computing architectures hello everyone and welcome to this week's wikibon cube insights powered by etr in this breaking analysis we'll explain why we believe nvidia is in the right position to power the world's computing centers and how it plans to disrupt the grip that x86 architectures have had on the data center for decades the data center market is in transition like the universe the cloud is expanding at an accelerated pace no longer is the cloud an opaque set of remote services i always say somewhere out there sitting in a mega data center no rather the cloud is extending to on-premises data centers data centers are moving into the cloud and they're connecting through adjacent locations that create hybrid interactions clouds are being meshed together across regions and eventually will stretch to the far edge this new definition or view of cloud will be hyper distributed and run by software kubernetes is changing the world of software development and enabling workloads to run anywhere open apis external applications expanding the digital supply chains and this expanding cloud they all increase the threat surface and vulnerability to the most sensitive information that resides within the data center and around the world zero trust has become a mandate we're also seeing ai being injected into every application and it's the technology area that we see with the most momentum coming out of the pandemic this new world will not be powered by general purpose x86 processors rather it will be supported by an ecosystem of arm-based providers in our opinion that are affecting an unprecedented increase in processor performance as we have been reporting and nvidia in our view is sitting in the poll position and is currently the favorite to dominate the next era of computing architecture for global data centers public clouds as well as the near and far edge let's talk about jensen wang's clarity of vision for this new world here's a chart that underscores some of the fundamental assumptions that he's leveraging to expand his market the first is that there's a lot of waste in the data center he claims that only half of the cpu cores deployed in the data center today actually support applications the other half are processing the infrastructure all around the applications that run the software defined data center and they're terribly under utilized nvidia's blue field three dpu the data processing unit was described in a blog post on siliconangle by analyst zias caravala as a complete mini server on a card i like that with software defined networking storage and security acceleration built in this product has the bandwidth and according to nvidia can replace 300 general purpose x86 cores jensen believes that every network chip will be intelligent programmable and capable of this type of acceleration to offload conventional cpus he believes that every server node will have this capability and enable every packed of every packet and every application to be monitored in real time all the time for intrusion and as servers move to the edge bluefield will be included as a core component in his view and this last statement by jensen is critical in our opinion he says ai is the most powerful force of our time whether you agree with that or not it's relevant because ai is everywhere an invidious position in ai and the architectures the company is building are the fundamental linchpin of its data center enterprise strategy so let's take a look at some etr spending data to see where ai fits on the priority list here's a set of data in a view that we often like to share the horizontal axis is market share or pervasiveness in the etr data but we want to call your attention to the vertical axis that's really really what really we want to pay attention today that's net score or spending momentum exiting the pandemic we've seen ai capture the number one position in the last two surveys and we think this dynamic will continue for quite some time as ai becomes the staple of digital transformations and automations an ai will be infused in every single dot you see on this chart nvidia's architectures it just so happens are tailor made for ai workloads and that is how it will enter these markets let's quantify what that means and lay out our view of how nvidia with the help of arm will go after the enterprise market here's some data from wikibon research that depicts the percent of worldwide spending on server infrastructure by workload type here are the key points first the market last year was around 78 billion dollars worldwide and is expected to approach 115 billion by the end of the decade this might even be a conservative figure and we've split the market into three broad workload categories the blue is ai and other related applications what david floyer calls matrix workloads the orange is general purpose think things like erp supply chain hcm collaboration basically oracle saps and microsoft work that's being supported today and of course many other software providers and the gray that's the area that jensen was referring to is about being wasted the offload work for networking and storage and all the software defined management in the data centers around the world okay you can see the squeeze that we think compute infrastructure is gonna gonna occur around that orange area that general-purpose workloads that we think is going to really get squeezed in the next several years on a percentage basis and on an absolute basis it's really not growing nearly as fast as the other two and video with arm in our view is well positioned to attack that blue area and the gray area those those workload offsets and the new emerging ai applications but even the orange as we've reported is under pressure as for example companies like aws and oracle they use arm-based designs to service general purpose workloads why are they doing that cost is the reason because x86 generally and intel specifically are not delivering the price performance and efficiency required to keep up with the demands to reduce data center costs and if intel doesn't respond which we believe it will but if it doesn't act arm we think will get 50 percent of the general purpose workloads by the end of the decade and with nvidia it will dominate the blue the ai and the gray the offload work when we say dominate we're talking like capture 90 percent of the available market if intel doesn't respond now intel they're not just going to sit back and let that happen pat gelsinger is well aware of this in moving intel to a new strategy but nvidia and arm are way ahead in the game in our view and as we've reported this is going to be a real challenge for intel to catch up now let's take a quick look at what nvidia is doing with relevant parts of its pretty massive portfolio here's a slide that shows nvidia's three chip strategy the company is shifting to arm-based architectures which we'll describe in more detail in a moment the slide shows at the top line nvidia's ampere architecture not to be confused with the company ampere computing nvidia is taking a gpu centric approach no surprise obvious reasons there that's their sort of stronghold but we think over time it may rethink this a little bit and lean more into npus the neural processing unit we look at what apple's doing what tesla are doing we see opportunities for companies like nvidia to really sort of go after that but we'll save that for another day nvidia has announced its grace cpu a nod to the famous computer scientist grace hopper grace is a new architecture that doesn't rely on x86 and much more efficiently uses memory resources we'll again describe this in more detail later and the bottom line there that roadmap line shows the bluefield dpu which we described is essentially a complete server on a card in this approach using arm will reduce the elapsed time to go from chip design to production by 50 we're talking about shaving years down to 18 months or less we don't have time to do a deep dive into nvidia's portfolio it's large but we want to share some things that we think are important and this next graphic is one of them this shows some of the details of nvidia's jetson architecture which is designed to accelerate those ai plus workloads that we showed earlier and the reason is that this is important in our view is because the same software supports from small to very large including edge systems and we think this type of architecture is very well suited for ai inference at the edge as well as core data center applications that use ai and as we've said before a lot of the action in ai is going to happen at the edge so this is a good example of leveraging an architecture across a wide spectrum of performance and cost now we want to take a moment to explain why the moved arm-based architectures is so critical to nvidia one of the biggest cost challenges for nvidia today is keeping the gpu utilized typical utilization of gpu is well below 20 percent here's why the left hand side of this chart shows essentially racks if you will of traditional compute and the bottlenecks that nvidia faces the processor and dram they're tied together in separate blocks imagine there are thousands thousands of cores in a rack and every time you need data that lives in another processor you have to send a request and go retrieve it it's very overhead intensive now technologies like rocky are designed to help but it doesn't solve the fundamental architectural bottleneck every gpu shown here also has its own dram and it has to communicate with the processors to get the data i.e they can't communicate with each other efficiently now the right hand side side shows where nvidia is headed start in the middle with system on chip socs cpus are packaged in with npus ipu's that's the image processing unit you know x dot dot dot x pu's the the alternative processors they're all connected with sram which is think of that as a high speed layer like an layer one cache the os for the system on a chip lives inside of this and that's where nvidia has this killer software model what they're doing is they're licensing the consumption of the operating system that's running this system on chip in this entire system and they're affecting a new and really compelling subscription model you know maybe they should just give away the chips and charge for the software like a razer blade model talk about disruptive now the outer layer is the the dpu and the shared dram and other resources like the ampere computing the company this time cpus ssds and other resources these are the processors that will manage the socs together this design is based on nvidia's three chip approach using bluefield dpu leveraging melanox that's the networking component the network enables shared dram across the cpus which will eventually be all arm based grace lives inside the system on a chip and also on the outside layers and of course the gpu lives inside the soc in a scaled-down version like for instance a rendering gpu and we show some gpus on the outer layer as well for ai workloads at least in the near term you know eventually we think they may reside solely in the system on chip but only time will tell okay so you as you can see nvidia is making some serious moves and by teaming up with arm and leaning into the arm ecosystem it plans to take the company to its next level so let's talk about how we think competition for the next era of compute stacks up here's that same xy graph that we love to show market share or pervasiveness on the horizontal tracking against next net score on the vertical net score again is spending velocity and we've cut the etr data to capture players that are that are big in compute and storage and networking we've plugged in a couple of the cloud players these are the guys that we feel are vying for data center leadership around compute aws is a very strong position we believe that more than half of its revenues comes from compute you know ec2 we're talking about more than 25 billion on a run rate basis that's huge the company designs its own silicon graviton 2 etc and is working with isvs to run general purpose workloads on arm-based graviton chips microsoft and google they're going to follow suit they're big consumers of compute they sell a lot but microsoft in particular you know they're likely to continue to work with oem partners to attack that on-prem data center opportunity but it's really intel that's the provider of compute to the likes of hpe and dell and cisco and the odms which are the odms are not shown here now hpe let's talk about them for a second they have architectures and i hate to bring it up but remember the machine i know it's the butt of many jokes especially from competitors it had been you know frankly hpe and hp they deserve some of that heat for all the fanfare and then that they they put out there and then quietly you know pulled the machine or put it out the pasture but hpe has a strong position in high performance computing and the work that it did on new computing architectures with the machine and shared memories that might be still kicking around somewhere inside of hp and could come in handy for some day in the future so hpe has some chops there plus hpe has been known hp historically has been known to design its own custom silicon so i would not count them out as an innovator in this race cisco is interesting because it not only has custom silicon designs but its entry into the compute business with ucs a decade ago was notable and they created a new way to think about integrating resources particularly compute and networking with partnerships to add in the storage piece initially it was within within emc prior to the dell acquisition but you know it continues with netapp and pure and others cisco invests they spend money investing in architectures and we expect the next generation of ucs oh ucs2 ucs 2.0 will mark another notable milestone in the company's data center business dell just had an amazing quarterly earnings report the company grew top line revenue by around 12 percent and it wasn't because of an easy compare to last year dells is simply executing despite continued softness in the legacy emc storage business laptop the laptop demand continued to soar in dell server business it's growing again but we don't see dell as an architectural innovator per se in compute rather we think the company will be content to partner with suppliers whether it's intel nvidia arm-based partners or all of the above dell we think will rely on its massive portfolio its excellent supply chain and execution ethos to compete now ibm is notable for historical reasons with its mainframe ibm created the first great compute monopoly before it unwind and wittingly handed it to intel along with microsoft we don't see ibm necessarily aspiring to retake that compute platform mantle that once once held with mainframes rather red hat in the march to hybrid cloud is the path that we think in our view is ibm's approach now let's get down to the elephants in the room intel nvidia and china inc china is of course relevant because of companies like alibaba and huawei and the chinese chinese government's desire to be self-sufficient in semiconductor technology and technology generally but our premise here is that the trends are favoring nvidia over intel in this picture because nvidia is making moves to further position itself for new workloads in the data center and compete for intel's stronghold intel is going to attempt to remake itself but it should have been doing this seven years ago what pat gelsinger is doing today intel is simply far behind and it's going to take at least a couple years for them to really start to to make inroads in this new model let's stay on the nvidia v intel comparison for a moment and take a snapshot of the two companies here's a quick chart that we put together with some basic kpis some of these figures are approximations or they're rounded so don't stress over it too much but you can see intel is an 80 billion dollar company 4x the size of nvidia but nvidia's market cap far exceeds that of intel why is that of course growth in our view it's justified due to that growth and nvidia's strategic positioning intel used to be the gross margin king but nvidia has much higher gross margins interesting now when it comes down to free cash flow intel is still dominant as it pertains to the balance sheet intel is way more capital intensive than nvidia and as it starts to build out its foundries that's going to eat into intel's cash position now what we did is we put together a little pro forma on the third column of nvidia plus arm circa let's say the end of 2022. we think they could get to a run rate that is about half the size of intel and that can propel the company's market cap to well over half a trillion dollars if they get any credit for arm they're paying 40 billion dollars for arm a company that's you know sub 2 billion the risk is that because of the arm because the arm deal is based on cash plus tons of stock it could put pressure on the market capitalization for some time arm has 90 percent gross margins because it pretty much has a pure license model so it helps the gross margin line a little bit for this in this pro forma and the balance sheet is a swag arm has said that it's not going to take on debt to do the transaction but we haven't had time to really dig into that and figure out how they're going to structure it so we took a took a swag in in what we would do with this low interest rate environment but but take that with a grain of salt we'll do more research in there the point is given the momentum and growth of nvidia its strategic position in ai is in its deep engineering they're aimed at all the right places and its potential to unlock huge value with arm on paper it looks like the horse to beat if it can execute all right let's wrap up here's a summary look the architectures on which nvidia is building its dominant ai business are evolving and nvidia is well positioned to drive a truck right to the enterprise in our view the power has shifted from intel to the arm ecosystem and nvidia is leaning in big time whereas intel it has to preserve its current business while recreating itself at the same time this is going to take a couple of years but intel potentially has the powerful backing of the us government too strategic to fail the wild card is will nvidia be successful in acquiring arm certain factions in the uk and eu are fighting the deal because they don't want the u.s dictating to whom arm can sell its technology for example the restrictions placed on huawei for many suppliers of arm-based chips based on u.s sanctions nvidia's competitors like broadcom qualcomm at all are nervous that if nvidia gets armed they will be at a competitive disadvantage they being invidious competitors and for sure china doesn't want nvidia controlling arm for obvious reasons and it will do what it can to block the deal and or put handcuffs on how business can be done in china we can see a scenario where the u.s government pressures the uk and eu regulators to let this deal go through look ai and semiconductors you can't get much more strategic than that for the u.s military and the u.s long-term competitiveness in exchange for maybe facilitating the deal the government pressures nvidia to guarantee some feed to the intel foundry business while at the same time imposing conditions that secure access to arm-based technology for nvidia's competitors and maybe as we've talked about before having them funnel business to intel's foundry actually we've talked about the us government enticing apple to do so but it could also entice nvidia's competitors to do so propping up intel's foundry business which is clearly starting from ground zero and is going to need help outside of intel's own semiconductor manufacturing internally look we don't have any inside information as to what's happening behind the scenes with the us government and so forth but on its earning call on its earnings call nvidia said they're working with regulators that are on track to complete the deal in early 2022. we'll see okay that's it for today thank you to david floyer who co-created this episode with me and remember i publish each week on wikibon.com and siliconangle.com these episodes they're all available as podcasts all you're going to do is search breaking analysis podcast and you can always connect with me on twitter at dvalante or email me at david.valante siliconangle.com i always appreciate the comments on linkedin and in the clubhouse please follow me so you can be notified when we start a room and riff on these topics and don't forget to check out etr.plus for all the survey data this is dave vellante for the cube insights powered by etr be well and we'll see you next time [Music] you

>>from around the globe. It's the Cube with digital coverage of biz ops Manifesto unveiled. Brought to you by biz ops Coalition. Hey, welcome back your body, Jeffrey here with the Cube. Welcome back to our ongoing coverage of the busy ops manifesto unveiling its been in the works for a while. But today is the day that it actually kind of come out to the to the public. And we're excited to have a real industry luminary here to talk about what's going on, Why this is important and share his perspective. And we're happy to have from Cape Cod, I believe, is Tom Davenport. He is a distinguished author on professor at Babson College. We could go on. He's got a lot of great titles and and really illuminate airy in the area of big data and analytics. Thomas, great to see you. >>Thanks, Jeff. Happy to be here with you. Great. >>So let's just jump into it, you know, and getting ready for this. I came across your LinkedIn post. I think you did earlier this summer in June and right off the bat, the first sentence just grabbed my attention. I'm always interested in new attempts to address long term issues, Uh, in how technology works within businesses. Biz ops. What did you see in biz ops? That that kind of addresses one of these really big long term problems? >>Well, yeah. The long term problem is that we've had a poor connection between business people and I t people between business objectives and the i t. Solutions that address them. This has been going on, I think, since the beginning of information technology, and sadly, it hasn't gone away. And so busy ops is new attempt to deal with that issue with a, you know, a new framework. Eventually a broad set of solutions that increase the likelihood that will actually solve a business problem with a nightie capability. >>Right. You know, it's interesting to compare it with, like, Dev ops, which I think a lot of people are probably familiar with, which was, you know, built around a agile software development and the theory that we want to embrace change that that changes okay on. We wanna be able to iterate quickly and incorporate that, and that's been happening in the software world for for 20 plus years. What's taking so long to get that to the business side because the pace of change is change on the software side. You know, that's a strategic issue in terms of execution on the business side that they need now to change priorities. And, you know, there's no P R D S and M R. D s and big giant strategic plans that sit on the shelf for five years. That's just not the way business works anymore. Took a long time to get here. >>Yeah, it did. And, you know, there have been previous attempts to make a better connection between business and i t. There was the so called strategic alignment framework that a couple of friends of mine from Boston University developed, I think more than 20 years ago. But, you know, now we have better technology for creating that linkage. And the, you know, the idea of kind of ops oriented frameworks is pretty pervasive now. So I think it's, um you know, time for another serious attempt at it, >>right? And do you think doing it this way right with the bizarre coalition, you know, getting a collection of of kind of like minded individuals and companies together and actually even having a manifesto which were making this declarative statement of principles and values. You think that's what it takes to kind of drive this, you know, kind of beyond the experiment and actually, you know, get it done and really start to see some results in, in in production in the field. >>I think certainly no one vendor organization can pull this off single handedly. It does require a number of organizations collaborating and working together. So I think a coalition is a good idea, and a manifesto is just a good way to kind of lay out. What you see is the key principles of the idea, and that makes it much easier for everybody. Toe I understand and act on. >>Yeah, I I think it's just it's really interesting having, you know, having them written down on paper and having it just be so clearly articulated both in terms of the of the values as well as as the the principles and and the values, you know. Business outcomes, matter, trust and collaboration, data driven decisions, which is the number three or four and then learn, responded pivot. It doesn't seem like those should have to be spelled out so clearly. But obviously it helps to have them there. You can stick them on the wall and kind of remember what your priorities are. But you're the data guy. You're the analytics guy. Yeah, And a big piece of this is data analytics and moving to data driven decisions. And principle number seven says, you know, today's organizations generate more data than humans can process. And informed decisions can be augmented by machine learning and artificial intelligence right up your alley. You know, you've talked a number of times on kind of the many stages of analytics. Onda. How has that's evolved over over time? You know, it is You think of analytics and machine learning driving decisions beyond supporting decisions, but actually starting to make decisions in machine time. What's that? What's that think for you? What does that make you? You know, start to think Wow, this is This is gonna be pretty significant. >>Yeah, well, you know, this has been a long term interest of mine. Um, the last generation of a I I was very interested in expert systems. And then e think more than 10 years ago, I wrote an article about automated decision making using, um, what was available then, which is rule based approaches. But, you know, this address is an issue that we've always had with analytics and ai. Um, you know, we tended Thio refer to those things as providing decision support. The problem is that if the decision maker didn't want their support, didn't want to use them in order to make a decision, they didn't provide any value. And so the nice thing about automating decisions with now contemporary ai tools is that we can ensure that data and analytics get brought into the decision without any possible disconnection. Now, I think humans still have something to add here, and we often will need to examine how that decision is being made and maybe even have the ability to override it. But in general, I think, at least for, you know, repetitive tactical decisions, um, involving a lot of data. We want most of those I think, to be at least, um, recommended, if not totally made by analgesic rhythm or an AI based system, and that, I believe would add to the quality and the precision and the accuracy of decisions. And in most organizations, >>you know, I think I think you just answered my next question before I Before I asked it. You know, we had Dr Robert Gates on the former secretary of Defense on a few years back, and we were talking about machines and machines making decisions, and he said at that time, you know, the only weapon systems that actually had an automated trigger on it, We're on the North Korean South Korea border. Um, everything else that you said had to go through some person before the final decision was made. And my question is, you know what are kind of the attributes of the decision that enable us that more easily automated? And then how do you see that kind of morphing over time both as the the data to support that as well as our comfort level, Um, enables us to turn mawr mawr actual decisions over to the machine? >>Well, yeah, I suggested we need data, and the data that we have to kind of train our models has to be high quality and current, and we need to know the outcomes of the that data. You know, most machine learning models, at least in business, are supervised, and that means we need tohave labeled outcomes in the in the training data. But you know, the pandemic that we're living through is a good illustration of the fact that the data also have to be reflective of current reality. And, you know, one of the things that were finding out quite frequently these days is that the data that we have a do not reflect you know what it's like to do business in a pandemic. I wrote a little piece about this recently with Jeff Cam at Wake Forest University. We call it Data Science Quarantined and it we interviewed somebody who said, You know, it's amazing what eight weeks of zeros will do to your demand forecast. We just don't really know what happens in a pandemic. Our models may be have to be put on the shelf for a little while and until we can develop some new ones or we can get some other guidelines into making decisions. So I think that's one of the key things with automated decision making. We have toe make sure that the data from the past and you know that's all we have, of course, is a good guide toe. You know what's happening in the present and in the future, as far as we understand it. >>Yeah, I used to joke when we started this calendar year 2020 was finally the year that we know everything with the benefit of hindsight. But it turned out 2020 the year we found out we actually know nothing and everything >>we thought we d >>o. But I wanna I wanna follow up on that because, you know, it did suddenly change everything, right? We got this light switch moment. Everybody's working from home now. We're many, many months into it, and it's going to continue for a while. I saw your interview with Bernard Marr and you had a really interesting comment that now we have to deal with this change. We don't have a lot of data and you talked about hold, fold or double down, and And I can't think of, um or, you know, kind of appropriate metaphor for driving the value of the biz ops. When now your whole portfolio strategy, um, needs to really be questioned. And, you know, you have to be really well executing on what you are holding. What you're folding and what you're doubling down with this completely new environment? >>Well, yeah, And I hope I did this in the interview. I would like to say that I came up with that term, but it actually came from a friend of mine was a senior executive at gen. Packed, and I used it mostly to talk about AI and AI applications, but I think you could You could use it much more broadly to talk about your entire sort of portfolio. Digital projects you need to think about. Well, um, given some constraints on resource is and a difficulty economy for a while. Which of our projects do we wanna keep going on Pretty much the way we were for and which ones, um, are not that necessary anymore. You see a lot of that in a I because we had so many pilots. Somebody told me, You know, we've got more pilots around here than O'Hare Airport in a I, um and then the the ones that involve double down there, even mawr Important to you, they are. You know, a lot of organizations have found this out in the pandemic on digital projects. It's more and more important for customers to be ableto interact with you digitally. And so you certainly wouldn't want toe cancel those projects or put them on hold. So you double down on them, get them done faster and better. >>Another. Another thing I came up in my research that that you quoted um, was was from Jeff. Bezos is talking about the great bulk of what we do is quietly but meaning fleeing, improving core operations. You know, I think that is so core to this concept of not AI and machine learning and kind of the general sense, which which gets way too much buzz but really applied, applied to a specific problem. And that's where you start to see the value. And, you know, the biz ops manifesto is calling it out in this particular process. But I just love to get your perspective. As you know, you speak generally about this topic all the time, but how people should really be thinking about where the applications where I can apply this technology to get direct business value. >>Yeah, well, you know, even talking about automated decisions, um, the kind of once in a lifetime decisions, uh, the ones that a G laugh. Li, the former CEO of Proctor and Gamble, used to call the big swing decisions. You only get a few of those, he said. In your tenure as CEO, those air probably not going to be the ones that you're automating in part because you don't have much data about them. Your you know, only making them a few times and in part because they really require that big picture thinking and the ability to kind of anticipate the future that the best human decision makers have. Um, but in general, I think where they I The projects that are working well are you know what I call the low hanging fruit ones? The some people even report to refer to it as boring A. I so you know, sucking data out of a contract in order to compare it Thio bill of lading for what arrived at your supply chain. Companies can save or make a lot of money with that kind of comparison. It's not the most exciting thing, but a I, as you suggest, is really good at those narrow kinds of tasks. Um, it's not so good at the at the really big Moonshots like curing cancer or, you know, figuring out well, what's the best stock or bond under all circumstances or even autonomous vehicles. We made some great progress in that area, but everybody seems to agree that they're not gonna be perfect for quite a while. And we really don't wanna be driving around on, um in that very much, unless they're, you know, good and all kinds of weather and with all kinds of pedestrian traffic. And you know that sort of thing, right? >>That's funny. Bring up contract management. I had a buddy years ago. They had a startup around contract management, and I'm like and this was way before we had the compute power today and cloud proliferation. I said, You know how How could you possibly built off around contract management? It's language. It's legalese. It's very specific. He's like Jeff. We just need to know where's the contract and when does it expire? And who's the signatory? And he built a business on those you know, very simple little facts that weren't being covered because their contracts from people's drawers and files and homes and Lord only knows so it's really interesting as you said. These kind of low hanging fruit opportunities where you could extract a lot of business value without trying to, you know, boil the ocean. >>Yeah, I mean, if you're Amazon, Jeff Bezos thinks it's important toe have some kind of billion dollar projects, and he even says it's important to have a billion dollar failure or two every year. But I think most organizations probably are better off being a little less aggressive and, you know, sticking to what a I has been doing for a long time, which is, you know, making smarter decisions based on based on data. >>Right? So, Tom, I want to shift gears one more time before before you let Ugo on on kind of a new topic for you, not really new, but you know, not not the vast majority of your publications. And that's the new way toe work, you know, as as the pandemic hit in mid March, right? And we had this light switch moment. Everybody had to work from home, and it was, you know, kind of crisis and get everybody set up. Well, you know, now we're five months, six months, seven months. A number of companies have said that people are not gonna be going back to work for a while, and so we're going to continue on this for a while, and then even when it's not what it is now, it's not gonna be what it was before. So, you know, I wonder and I know you, you tease. You're working on a a new book, you know, some of your thoughts on, you know, kind of this new way, uh, toe work and and and the human factors in this new, this new kind of reality that we're kind of evolving into, I guess, >>Yeah, this was an interest of mine. I think. Back in the nineties, I wrote an article called a co authored an article called Two Cheers for the Virtual Office. And, you know, it was just starting to emerge than some people were very excited about it. Some people were skeptical, and we said to cheers rather than three cheers because clearly there's some shortcomings and, you know, I keep seeing these pop up. It's it's great that we can work from our homes. It's great that we can accomplish most of what we need to do with a digital interface, but you know, things like innovation and creativity and certainly, um a A good, um, happy social life kind of requires some face to face contact every now and then. And so you know, I think we'll go back to an environment where there is some of that. Um, will have, um, time when people convene in one place so they can get to know each other face to face and learn from each other that way. And most of the time, I think it's a huge waste of people's time to commute into the office every day and toe jump on airplanes. Thio, Thio, give every little sales call or give every little presentation we just have to really narrow down. What are the circumstances, where face to face contact really matters and when can we get by with digital? You know, I think one of the things in my current work on finding is that even when you have AI based decision making, you really need a good platform in which that all takes place. So in addition to these virtual platforms, we need to develop platforms that kind of structure the workflow for us and tell us what we should be doing next and make automated decisions when necessary. And I think that ultimately is a big part of biz ops as well. It's not just the intelligence of an AI system, but it's the flow of work that kind of keeps things moving smoothly throughout your organization. Yeah, >>I think such such a huge opportunity as you just said, because I forget the stats on how often were interrupted with notifications between email text, slack asana, salesforce The list goes on and on. So, you know, t put an AI layer between the person and all these systems that are begging for attention. And you've written a you know, a book on the attention economy, which is a whole nother topic will say for another day. You know, it really begs. It really begs for some assistance because, you know, you just can't get him picked, you know, every two minutes and really get quality work done. It's just not it's just not realistic. And you know what? I don't think that's the future that we're looking for. >>Great. Totally. Alright, >>Tom. Well, thank you so much for your time. Really enjoyed the conversation. I got to dig into the library. It's very long song. I might started the attention economy. I haven't read that one in to me. I think that's the fascinating thing in which we're living. So thank you for your time. And, uh, great to see you. >>My pleasure, Jeff. Great to be here. >>All right, take care. Alright. East, Tom. I'm Jeff. You are watching the continuing coverage of the biz ops manifesto. Unveil. Thanks for watching the Cube. We'll see you next time.

>>from around the globe. It's the Cube with digital coverage of biz ops Manifesto unveiled. Brought to you by biz ops Coalition. Hey, welcome back your body, Jeffrey here with the Cube. Welcome back to our ongoing coverage of the busy ops manifesto unveiling its been in the works for a while. But today is the day that it actually kind of come out to the to the public. And we're excited to have a real industry luminary here to talk about what's going on, Why this is important and share his perspective. And we're happy to have from Cape Cod, I believe, is Tom Davenport. He is a distinguished author on professor at Babson College. We could go on. He's got a lot of great titles and and really illuminate airy in the area of big data and analytics. Thomas, great to see you. >>Thanks, Jeff. Happy to be here with you. Great. >>So let's just jump into it, you know, and getting ready for this. I came across your LinkedIn post. I think you did earlier this summer in June and right off the bat, the first sentence just grabbed my attention. I'm always interested in new attempts to address long term issues, Uh, in how technology works within businesses. Biz ops. What did you see in biz ops? That that kind of addresses one of these really big long term problems? >>Well, yeah. The long term problem is that we've had a poor connection between business people and I t people between business objectives and the i t. Solutions that address them. This has been going on, I think, since the beginning of information technology, and sadly, it hasn't gone away. And so busy ops is new attempt to deal with that issue with a, you know, a new framework. Eventually a broad set of solutions that increase the likelihood that will actually solve a business problem with a nightie capability. >>Right. You know, it's interesting to compare it with, like, Dev ops, which I think a lot of people are probably familiar with, which was, you know, built around a agile software development and the theory that we want to embrace change that that changes okay on. We wanna be able to iterate quickly and incorporate that, and that's been happening in the software world for for 20 plus years. What's taking so long to get that to the business side because the pace of change is change on the software side. You know, that's a strategic issue in terms of execution on the business side that they need now to change priorities. And, you know, there's no P R D S and M R. D s and big giant strategic plans that sit on the shelf for five years. That's just not the way business works anymore. Took a long time to get here. >>Yeah, it did. And, you know, there have been previous attempts to make a better connection between business and i t. There was the so called strategic alignment framework that a couple of friends of mine from Boston University developed, I think more than 20 years ago. But, you know, now we have better technology for creating that linkage. And the, you know, the idea of kind of ops oriented frameworks is pretty pervasive now. So I think it's, um you know, time for another serious attempt at it, right? >>And do you think doing it this way right with the bizarre coalition, you know, getting a collection of of kind of like minded individuals and companies together and actually even having a manifesto which were making this declarative statement of principles and values. You think that's what it takes to kind of drive this, you know, kind of beyond the experiment and actually, you know, get it done and really start to see some results in, in in production in the field. >>Well, you know, the manifesto approach worked for Karl Marx and communism. So maybe it'll work. Here is Well, now, I think certainly no one vendor organization can pull this off single handedly. It does require a number of organizations collaborating and working together. So I think a coalition is a good idea, and a manifesto is just a good way to kind of lay out. What you see is the key principles of the idea, and that makes it much easier for everybody. Toe I understand and act on. >>Yeah, I I think it's just it's really interesting having you know, having them written down on paper and having it just be so clearly articulated both in terms of the of the values as well as as the the principles and and the values, you know, business outcomes, matter, trust and collaboration, data driven decisions, which is the number three or four and then learn responded Pivot, It doesn't seem like those should have to be spelled out so clearly, but obviously it helps to have them there. You can stick them on the wall and kind of remember what your priorities are. But you're the data guy. You're the analytics guy. Uh, and a big piece of this is data analytics and moving to data driven decisions. And principle number seven says, you know, today's organizations generate more data than humans can process. And informed decisions can be augmented by machine learning and artificial intelligence right up your alley. You know, you've talked a number of times on kind of the many stages of analytics Onda how that's evolved over over time. You know, it is you think of analytics and machine learning driving decisions beyond supporting decisions, but actually starting to make decisions in machine time. What's that? What's that think for you? What does that make you? You know, start to think Wow, this is this is gonna be pretty significant. >>Yeah, well, you know, this has been a long term interest of mine. Um, the last generation of a I I was very interested in expert systems. And then e think more than 10 years ago I wrote an article about automated decision making using, um, what was available then, which is rule based approaches. But, you know, this address is an issue that we've always had with analytics and ai. Um, you know, we tended Thio refer to those things as providing decision support. The problem is that if the decision maker didn't want their support, didn't want to use them in order to make a decision, they didn't provide any value. And so the nice thing about automating decisions with now contemporary ai tools is that we can ensure that data and analytics get brought into the decision without any possible disconnection. Now, I think humans still have something to add here, and we often will need to examine how that decision is being made and maybe even have the ability to override it. But in general, I think, at least for, you know, repetitive tactical decisions, um, involving a lot of data. We want most of those I think, to be at least, um, recommended, if not totally made by analgesic rhythm or an AI based system, and that I believe would add to the quality and the precision and the accuracy of decisions in in most organizations. >>You know, I think I think you just answered my next question before I before I asked it. You know, we had Dr Robert Gates on the former secretary of Defense on a few years back, and we were talking about machines and machines making decisions, and he said at that time, you know, the only weapon systems that actually had an automated trigger on it, We're on the North Korea and South Korea border. Everything else, as you said, had to go through some person before the final decision was made. And my question is, you know what are kind of the attributes of the decision that enable us to more easily automated? And then how do you see that kind of morphing over time both as the data to support that as well as our comfort level, Um, enables us to turn Maura Maura actual decisions over to the machine? >>Well, yeah, I suggested we need data and the data that we have to kind of train our models has to be high quality and current, and we need to know the outcomes of that data. You know, most machine learning models, at least in business, are supervised, and that means we need tohave labeled outcomes in the in the training data. But, you know, the pandemic that we're living through is a good illustration of the fact that the the data also have to be reflective of current reality. And, you know, one of the things that we're finding out quite frequently these days is that the data that we have do not reflect. You know what it's like to do business in it. Pandemic it. I wrote a little piece about this recently with Jeff Cam at Wake Forest University. We call it Data Science quarantined, and we interviewed somebody who said, You know, it's amazing what eight weeks of zeros will do to your demand forecast. We just don't really know what happens in a pandemic. Our models may be have to be put on the shelf for a little while and until we can develop some new ones or we can get some other guidelines into making decisions. So I think that's one of the key things with automated decision making. We have toe, make sure that the data from the past and you know, that's all we have, of course, is a good guide toe. You know what's happening in the present and and the future as far as we understand it. >>Yeah, I used to joke when we started this calendar year 2020 is finally the year that we know everything with the benefit of hindsight. But it turned out 2020 the year we found out we actually know nothing and everything way. But I wanna I wanna follow up on that because, you know, it did suddenly change everything, right? We got this light switch moment. Everybody's working from home now. We're many, many months into it, and it's going to continue for a while. I saw your interview with Bernard Marr and you had a really interesting comment that now we have to deal with this change. We don't have a lot of data and you talked about hold, fold or double down and and I can't think of, um or, you know, kind of appropriate metaphor for driving the value of the biz ops. When now your whole portfolio strategy, um, needs to really be questioned. And, you know, You have to be really well, executing on what you are holding, what you're folding and what you're doubling down with this completely new environment. >>Well, yeah, And I hope I did this in the interview. I would like to say that I came up with that term, but it actually came from a friend of mine who's a senior executive at gen. Packed. And I used it mostly to talk about AI and AI applications, but I think you could You could use it much more broadly to talk about your entire sort of portfolio of digital projects you need to think about. Well, um, given some constraints on resource is and a difficulty economy for a while. Which of our projects do we wanna keep going on Pretty much the way we were And which ones, um, are not that necessary anymore. You see a lot of that in a I because we had so many pilots, somebody for me, you know, we've got more pilots around here, then O'Hare airport in a I, um and then the the ones that involve double down there, even mawr Important to you, they are, you know, a lot of organizations have found this out in the pandemic on digital projects, it's more and more important for customers to be ableto interact with you, um, digitally. And so you certainly wouldn't want toe cancel those projects or put them on hold. So you double down on them, get them done faster and better. >>Another. Another thing that came up in my research that that you quoted, um, was was from Jeff. Bezos is talking about the great bulk of what we do is quietly but meaning fleeing, improving core operations. You know, I think that is so core to this concept of not AI and machine learning and kind of the general sense, which which gets way too much buzz but really applied, applied to a specific problem. And that's where you start to see the value and, you know, the biz ops. Uh, manifesto is calling it out in this particular process, but I just love to get your perspective. As you know, you speak generally about this topic all the time, but how people should really be thinking about where the applications where I can apply this technology to get direct business value. >>Yeah, well, you know, even talking about automated decisions? Uh, the kind of once in a lifetime decisions, uh, the ones that a g laugh Li, the former CEO of Proctor and Gamble, used to call the big swing decisions. You only get a few of those, he said. In your tenure as CEO, those air probably not going to be the ones that you're automating in part because you don't have much data about them. You're only making them a few times, and in part because they really require that big picture thinking and the ability to kind of anticipate the future that the best human decision makers have. Um, but in general, I think where they I the projects that are working well are you know what I call the low hanging fruit ones? The some people even report to refer to it as boring A I so you know, sucking data out of a contract in order to compare it Thio bill of lading for what arrived at your supply chain. Companies can save or make a lot of money with that kind of comparison. It's not the most exciting thing, but a I, as you suggest, is really good at those narrow kinds of tasks. Um, it's not so good at the at the really big Moonshots like curing cancer or, you know, figuring out well, what's the best stock or bond under all circumstances or even autonomous vehicles. We made some great progress in that area, but everybody seems to agree that they're not going to be perfect for quite a while. And we really don't wanna be driving around on, um in that very much, unless they're, you know, good and all kinds of weather and with all kinds of pedestrian traffic. And you know that sort of thing, right? >>That's funny. Bring up contract management. I had a buddy years ago. They had a startup around contract management, and I'm like, and this was way before we had the compute power today and and cloud proliferation. I said, You know how How could you possibly built off around contract management? It's language. It's legalese. It's very specific. He's like Jeff. We just need to know where's the contract and when does it expire? And who's a signatory? And he built a business on those you know, very simple little facts that weren't being covered because their contracts from People's drawers and files and homes, and Lord only knows So it's really interesting, as you said, these kind of low hanging fruit opportunities where you could extract a lot of business value without trying to, you know, boil the ocean. >>Yeah, I mean, if you're Amazon, Jeff Bezos thinks it's important toe have some kind of billion dollar projects, and he even says it's important to have a billion dollar failure or two every year. But I think most organizations probably are better off being a little less aggressive and, you know, sticking to what a I has been doing for a long time, which is, you know, making smarter decisions based on based on data. >>Right? So, Tom, I want to shift gears one more time before before you let Ugo on on kind of a new topic for you, not really new, but you know, not not the vast majority of your publications. And that's the new way toe work, you know, as as the pandemic hit in mid March, right? And we had this light switch moment. Everybody had to work from home, and it was, you know, kind of crisis and get everybody set up well you know, Now we're five months, six months, seven months. A number of companies have said that people are not gonna be going back to work for a while. And so we're going to continue on this for a while, and then even when it's not what it is now, it's not gonna be what it was before. So, you know, I wonder and I know you, you tease. You're working on a a new book, you know, some of your thoughts on, you know, kind of this new way. Uh, toe work and and and the human factors in this new, this new kind of reality that we're kind of evolving into, I guess. >>Yeah, This was an interest of mine. I think back in the nineties, I wrote an article called Ah Co authored an article called Two Cheers for the Virtual Office. And, you know, it was just starting to emerge. Then some people were very excited about it. Some people were skeptical and we said to cheers rather than three cheers because clearly there's some shortcomings and, you know, I keep seeing these pop up. It's great that we can work from our homes. It's great that we can accomplish most of what we need to do with a digital interface. But you know, things like innovation and creativity and certainly a a good, um, happy social life kind of requires some face to face contact every now and then. And so you know, I think we'll go back to an environment where there is some of that. We'll have, um, time when people convene in one place so they can get to know each other face to face and learn from each other that way. And most of the time, I think it's a huge waste of people's time to commute into the office every day and toe jump on airplanes. Thio, Thio give every little mhm, uh, sales call or give every little presentation. We just have to really narrow down. What are the circumstances, where face to face contact really matters and when can we get by with digital? You know, I think one of the things in my current work I'm finding is that even when you have a I based decision making, you really need a good platform in which that all takes place. So in addition to these virtual platforms, We need to develop platforms that kind of structure the workflow for us and tell us what we should be doing next and make automated decisions when necessary. And I think that ultimately is a big part of biz ops as well. It's not just the intelligence oven, a isis some, but it's the flow of work that kind of keeps things moving smoothly throughout your organization. Yeah, >>I think such such a huge opportunity as you just said, because I forget the stats on how often were interrupted with notifications between email text, slack asana, salesforce The list goes on on and on. So, you know, t put an AI layer between the person and all these systems that are begging for attention. And you've written a you know, a book on the attention economy, which is a whole nother topic will say for another day. You know, it really begs. It really begs for some assistance because, you know, you just can't get him picked, you know, every two minutes and really get quality work done. It's just not it's just not realistic. And you know what? I don't think that's the future that we're looking for. >>Great totally alright, >>Tom. Well, thank you so much for your time. Really enjoyed the conversation. I gotta dig into the library. It's very long song. I might started the attention economy. I haven't read that one in to me. I think that's the fascinating thing in which we're living. So thank you for your time. And, uh, great to see you. >>My pleasure, Jeff. Great to be here. >>All right, take care. Alright. He's Tom. I'm Jeff. You are watching the continuing coverage of the biz ops manifesto. Unveil. Thanks for watching. The Cube will see you next time.

>> from Burlingame, California It's the Cube covering Suma logic Illuminate 2019. Brought to You by Sumer Logic >> Hey, welcome back, everybody. Jeffrey here with the Cube were at the higher Regency San Francisco Airport at Suma Logic, Illuminate, 2019 were here last year for our first time. It's a 30 year the show. It's probably 809 100 people around. 1000 packed house just had the finish. The keynote. And we're really excited to have our first guest of the day. Who's been here since the very beginning is Bruno Critic, the founding VP of product and strategy for Suma Logic, you know, great to see you. Likewise. Thank you. So I did a little homework and you're actually on the cube aws reinvent, I think 2013. Wow. How far has the cloud journey progressed? Since efforts? I think it was our first year at reinvented as well. >> That's the second year agreement, >> right? So what? What an adventure. You guys made a good bet six years ago. Seems to be paying off pretty well. >> It really has been re kind of slipped out that the cloud is gonna be a real thing. Put all of our bats into it and have been executing ever since. And I think we were right. They think it is no longer a question. Is this cloud thing gonna be re alarm enterprise gonna adopt it? It's just how quickly and how much. >> Right? Right. But we've seen kind of this continual evolution, right? Was this jump into public cloud? Everybody jumped in with both feet, and now they're pulling back a little bit. But now really seen this growth of the hybrid cloud Big announcement here with Antos and Google Cloud Platform and in containers. And, you know, the rise of doctor and the rise of kubernetes. So I don't know, a CZ. You look a kind of the evolution. A lot of positive things kind of being added to the ecosystem that have helped you guys in your core mission. >> That's right. Look, you know, five years ago, which is such a short time, But yet instead of the speed of the technology adoption and change, you know it's in It's in millennia. What's happened over the last few years is technology stocks have changed dramatically. We've gone from okay, we can host some v ems in the cloud and put some databases in the cloud. So we're now building micro service's architecture, leveraging new technologies like Kubernetes like Serverless Technologies and all the stuff And, you know, some one of the fastest growing technologies that's being adopted by some village custom base, actually the fastest kubernetes and also the fastest customer segment growing customer segments. ImmuLogic is multi clog customers, basically that sort of desire by enterprise to build choice into their offerings. Being able to have leverage over the providers is really coming to fruition right now, >> right? But the multi cloud almost it makes a lot of sense, right, because we're over and over. You want to put your workload in the environment that supposed appropriate for the workload. It kind of. It kind of flipped the bid. It was no longer. Here's your infrastructure. What kind of APs can you build on it? Now here's my app. Where should it run that maybe on Prem it may be in a public cloud. It may be in a data center, so it's kind of logical that we've come into this this hybrid cloud world that said, Now you've got a whole another layer of complexity that that's been added on. And that's really been a big part of the rise of kubernetes. >> That's right. And so, as you're adopting service's that are not equal, right, you have to create a layer that insulate you from those. Service is if you look a tw r continues intelligence report that we just announced today. You will also see that how customers and enterprise are adopting cloud service is is they're essentially adopting the basic and core compute storage network, and database service is there's a long, long tail of service that are very infrequently adopted. And that is because enterprise they're looking for a way to not get to lock Tintin into anyone. Service provider kubernetes Give them Give them that layer of insulation with in thoughts and other technologies like that, you are now able to seamlessly manage all those workloads rather there on your on premise in AWS in G C. P. In azure or anywhere else, >> right? So there's so much we can unpack. You're one of the things I want to touch on which you talked about six years ago, but it's even more thing appropriate. Today is kind of this scale this exponential growth of data on this exponential scale of complexity. And we, as people, has been written about by a lot of smart people, and I, we have a real hard time. Is humans with exponential growth. Everything's linear. Tow us. So as you look at this exponential growth and now we're trying to get insights. Now we've got a I ot and this machine a machine data, which is a whole another multiple orders of magnitude. You can't work in that world with a single painted glass with somebody looking at a dashboard that's trying to find a yellow light that's earned it. I'm going to go read. You don't have analytics. Your hose. >> That's right. This is no longer world of Ding dong lights, right? You can just like to say, Okay, red, green, yellow. The as sort of companies go digital right? Which is driving this growth in data, you know? Ultimately, that data is governed by Moore's law. Moore's law says machines are gonna be able to do twice as much every 18 to 24 months. Well, that guess what? They're gonna tell you what they're doing twice as much. Every 18 to 24 months, and that is an exponential growth rate, right? The challenge that is, budgets don't grow at that rate, either, right? So budgets are not exponentially growing. So how do you cope with the onslaught of this data? And if you're running a digital service, right, if you're serving your customers digital generating revenue through digital means, which is just about every industry. At this point in time, you must get that data because if you don't get the data, you can't run your business. This data is useful not just in operations and security. It's useful for general business abuse, useful in marketing and product management in sales and their complexity. And the analytics required to actually make sense of that data and serve it to the right constituency in the business is really hard. And that has been whatever we have been trying to solve, including this economics of machine. Dad and me talked about it today. Keynote. We're trying t bend the cost curve >> Moore's law >> yet delivered analytics that the enterprise can leverage to really not just operate an application but run their business >> right. So let's talk about this concept of observe ability. You've written box about it. When you talk to people about observe ability, what should they be thinking about? How are you defining it? Why is it important? >> It's great question, So observe ability right now is being defined as a technique right. The simplest way to think about it is people think, observe a witty I need to have these three data sets and I have observed ability. And then you have to ask yourself a question. First of all, what is Observe ability and why does it matter? I think there's a a big misconception in the market how people adopt this is that they think, observe abilities the end. But it isn't observe. Ability is the means of achieving a goal. And what we like to talk about is what is the goal? Observe, observe ability right now. Observe abilities talked about strictly in the devil up space, right? Basically, how am I going to get obs Erv City into an application? And it's maybe runtime how it's running, whether it's up and performance. The challenge with that is that is a pigeon pigeon hole view off, observe ability, observe ability. If you think about it, we talk about objectives during observe ability. Operability tau sa two ns Sorry could be up time in performance. Well, guess what a different group like security observe. Ability is not getting breached. Understanding your compliance posture. Making sure that you are compliant with with regular to re rules and things like that observe ability to a business person to a product manager who's who owns a P N. L. On some product is how are my users using this product powers my application being adopted where users having trouble. What are they and where's the user experience? Poor right? So all of this data is multifaceted and multi useful as multi uses and observing Tow us. Is his objectives driven? If you don't know what your object it is, observe. Ability is just a tool. >> I love that, you know, because it falls under this thing We talked about off the two, which is, you know, there's data, right, and then there's information in the data and then, but it is a useful information because it has to be applied to something that's right in and of itself. It has no value, and what you're talking about really is getting the right data to the right person at the right time, which kind of stumbled into another area, which is how do you drive innovation in an organization? In one of the simple concepts is democratization. Get more people more than data more than tools to manipulate the data. Then piano manager is gonna make a different decision based on different visibility than Security Person or the Dev Ops person. So how is how is that evolving? Where do you see it going? Where was it in the past? And you know, I think he made it interesting or remain made. Interesting thing in the keynote where you guys let your software be available to everyone. And there was a lot of people talking about giving Maur. People Maur access to the tools and more of the data so that they can start to drive this innovation >> abuse of an example of one of the one of the sort of aspects of when we talk about continued continues intelligence. What do we mean? So this concept of agile development didn't evolve because people somehow thought, Hey, why don't we just try to push court production all the time? Break stuff all the time. What's the What's the reason why that came about? It did not come about because somehow somebody decided so better. Software development model It's because cos try to innovate faster, so they they wanted Toa accelerate. How they deliver digital product and service is to their customers. And what's facilitates that delivery cycle is the feedback loop. They get out of their data. They push code early. They observed the data. They understand what it's telling them about how their customers are using their products, and service is what products are working with or not. And they're quickly baking that feedback back into their development cycles into the business business cycles. To make better Prada effectively, it evolved as a as a tool to differentiate and out innovate the competition. And that's to a large degree one of the ways that you deliver the right inside to the right group to improve your business right. And so this is applicable across all use cases in order pot. All departments are on the company, but that's just one example of how you think of this continuous innovation, continuous data from to use analytics and don't >> spend two years doing an M r d and another two years doing a P R d and then another to your shift >> When you when you actually ship it. Half of the assumptions that you made two years ago already all the main along, right? So now you've gotta go. You've wasted half of your development time, and you've only released half of the value that you could have other, >> right? Right. And your assumptions are not gonna be correct, right? You just don't know until you get that >> you think over time, like two years of kubernetes with a single digits percentage adoption technology and soon was customer base. Now it's 1/3 right? Right? Which means no things have changed. If I had made an assumption as of two years ago on communities, I would have no way wouldn't have done this announcement, >> right? Right. >> But we did it in an interactive mode and re benefit from that continuous information continues intelligence that we do in our own >> right, right? We fed Joe and the boys on lots of times so that it's a pretty interesting how fast that came and how it really kind of over took. Doctor has informed they contain it. Even the doctor, according to reporters. Still getting a Tana Tana traction >> and it's >> working in conjunction with communities. Communities allows you to manage those containers right, And Dr Containers are always part of the ecosystem. And so it's, you know, you know, it's like the management layer and the actual container layer, >> right? So as you look forward to give you the last word, you know, as we're really kind of getting into the SIA Teague World and five G's coming just around around the corner, which is gonna have a giant impact on an industrial I ity and this machine a machine communications, what are some of your priorities? What are you looking, you know, kind of a little bit down the road and keeping an eye on >> interesting question. You know, we used to think about I ot as is the new domain. We should think about I or tea. And maybe we need to build a solution for right. It turns out our biggest customers, customers and the way that I have personally reframed my thinking about Iris is the following Computational capacity is ubiquitous. Now, what used to be a modern application 345 years ago was something that your access to your laptop or three or mobile app, and maybe you're a smart watch Now the computation that you interface with runs in your doorbell, you know, in a light switch in your light bulbs and how's it runs everywhere runs in your shoe because when you're around, it talks to your phone to tell you how many steps you've taken, all the stuff right? Essentially, enterprises building application to serve their customers are simply pushing computation farther and farther into our being, like everywhere. There's now I, P Networks, CP use memory and all of those distributed computers are now running the applications that are serving us in our lives, right? And to me, that's what I ot is. It's just an extension off what the digital service is our and we interface with does, and it so happens that when you push computation farther and farther into our lives, you get more and more computers participating. You get more data, and many of our largest customers are essentially ingesting their full stack of iron devices to serve their customers >> right crazy future and you know, it just kind of this continual Adam ization to of computer store and memory. Well, Bruno, hopefully it will not be six years before we see you again. Congrats on the conference. And thanks for taking a few minutes. Absolutely. All right. He's Bruno. I'm Jeff. You're watching the Cube where? It's suma logic illuminate at the Hyatt Regency seven square port. Thanks for watching. We'll see you next time.

(energetic music) >> Announcer: Live from Las Vegas, it's the CUBE, covering IBM Think 2018. Brought to you by IBM. >> Hello everyone, welcome back to the cube's coverage here at IBM Think 2018. We are in Las Vegas, the Mandalay Bay, for IBM Think. Six shows are coming into one packed house. We have two great guests here, Brian Wallace, who's the CTO of Insurance for DXC technologies, and we have, Bala, The Bala, but goes by Bala, banking and capital markets CTO for DXC technologies. Guys, welcome to the cube. Thanks for joining us. >> Thank you. >> It's our pleasure, yeah, thanks. >> So, the innovation sandwich, I'm calling it IBM strategy. You've got in the middle, the meat, is data. And the bread is blockchain and AI. Two really fundamental technologies powered by cloud and a variety of other things. Obviously, AI is disrupted, we know what that looks like. Block Chain now emerging as a viable infrastructure enabler that's creating token economics, a lot of cool things, certainly on the banking side, seeing a lot of controversy. Block Chain really is driving it. You guys are out on the front lines. You're doing a lot crowd chats, been following your digital transformation story that you guys have been putting out there. Really you're on this. So, what's the conversations like that you guys are having with block chain and AI; Share? >> Bala: So, let me begin with a couple of quick points on block chain. DXC has done some fantastic work around the world leveraging both the trust capability that block chain brings to bear in financial banking industry use cases, like KYC for instance, institutional KYC in particular, but also, in simplification of entire value chains such as lending. And we're doing very interesting work in lending where not only are we looking at the up-front origination process of lending but also the downstream securitization. Which is where the tokenization of principle and interest payments and those type of things happen. >> John: Energy too? >> Oh yes, absolutely. So there are a number of these creating type use cases that follow into securitization. And with that, we're doing some very interesting work. >> John: Bala, talk about the globalization because one of the things we're seeing in the US a shrinking middle class, but outside the US in emerging markets, a growing middle class. Thanks to mobile technology, thanks to data, thanks to block chain, you're seeing, you know, countries that "hey, we have infrastructure but we don't have the core and modern infrastructure but you throw in a decentralized capability, You've got all these capabilities, and the killer app in all this is money. You're in, that's your vertical. >> Bala: Yes. >> That's your industry. The killer app is money and marketplaces. Your thoughts? >> Bala: I think, the beauty of what these technologies are doing, is for the first time creating financial inclusion to happen and the very first case of where financial inclusion is enabled, is in payments. So, when we open up the banking system predominantly from a payment perspective, which is what things like blockchain and others enable, if we succeed in doing that, then for the first time we've enabled, that's 2 billion people unbanked or underbanked-2 billion. >> John: Yeah. >> Bringing them into this financial system allows for. >> And some people are discriminated against too because they don't have a track record. Banks can't handle some of the things that others are now filling the void with crypto and blockchain. >> Bala: Right, or they can't service them profitably. But for the first time now, you're looking at the economics that cloud, and AI, and blockchain, these technologies bring, not just into banking and capital markets areas but into insurance and I'd love to have my colleague, Brian, talk with the insurance cases are enabled as well. >> John: Brian, insurance- go. >> Yeah, so it's a slightly different dynamic. There it's the, if you think about the fundamental pattern of blockchain it's around eliminating a central or a middle-man or a central, you know, gatekeeper, if you will. And the entire insurance industry is largely made up of middle-men, right? You've got people with risk at one end and you've got sources of capital at the other end and everybody's playing a role between a broker, and a carrier, and a re-insurer. In sort of facilitating that management and that transfer of risk. >> John: So you've got to extract some efficiencies out of that. Business model opportunity. >> So efficiencies, there's a lot of conversations around efficiencies, around automation, but interestingly, it's around the disruptive business model, right? The technology is mildly interesting but it's the new business models that blockchain will enable. >> John: Yeah, I see banking picking up. The early adopter on blockchain but I see, maybe it lagging a bit in insurance but I definitely see some opportunity there. But short term, data is driving insurance because, you know, I don't have a Tesla but my friend has a Tesla. The insurance company will know exactly who is rolling through those stop signs. They know everything that he's doing, All the data is there, so AI becomes really the low hanging fruit for insurance in that industry. Do you agree with that? Comment, reaction? >> Brian: Yeah, and we're just at the beginning, right? Because as you say, data is the asset that we manage. So we have a lot of data in terms of transactional data, the traditional operational data. What we're discovering, and what we're sort of licking our lips over almost is all of this new unstructured data, whether it's sensor data, behavioral data, and you're right, 'cause the challenge that we had around automation and cognitive computing, if you will. We're here at IBM with the Watson tech, was enough data, and the consistency and quality of that data. So we have that now, and we're making tremendous strides around in particular here, with the Watson brand, and the Watson cognitive. >> John: You know, one of the things I wish, was Dan Hutches was here, he's not, he's the CTO in charge. You've guys have been doing all these crowd chats our software that we wrote. That's pretty interesting. I've personally enjoyed all the conversations and give a shout out to Dan and you guys for really great conversation. You guys know what you're talking about. It's clear in the data you guys are taking an outside-in approach and collaborating. But your topics are on target. You're talking about digital transformation kind of holistically, but then you start to dive down into specific use cases. So, Bala, what is the favorite, or the most popular digital disruptive topic that's being discussed within DXC and your clients and in the marketplace? >> So, at the outset, within DXC, as digital transformation takes hold with our customers and we aim to be the premier provider of that enablement, what we've realized ourselves is that we provide a lot of services to our clients across many industries but there are commonalities across what we provide in terms of service delivery. And so it made sense for us to, number one: look at the commonalities and create a platform that was common across industries, across offerings that we bring to the marketplace. That commonality is what we call internally, and externally now, as bionics. And it's a platform that we are bringing forward that for the first time ties together what we are talking about both here at this event but also with our clients. Ties together intelligence, orchestration, and automation which are the fundamental, >> John: It's called bionics? >> Bionics. And internally we call it platform DXC upon which all of our offerings and services are brought to market. >> John: Well there's disruption going on in your business. So, I want to talk about, double-down on that for a second. I'm seeing a trend, certainly in the public sector market where the use cases are well enough defined. So you're seeing automatic code generation becoming a real part of the delivery process. Now, what that's going to do is essentially, think of provisioning and configuration management in cloud. If you could apply actual process code that you've done before in the commonalities, this is going to change the delivery timeframe. So you're looking at essentially auto-provisioning software. Not just like, configuration management resources. No, I'm saying here's a value chain, here's a block chain, here's some AI, just configure it like a LEGO block, push. That could take months to deliver the old way. >> Bala: Right. >> Your thoughts to that? Are you guys on that? Do you guys see that as something that's going to be an opportunity for you? Some companies, I've seen, Global system integrator, is being disrupted by this, cause they don't have this. New SI's, new system integrators, are thinking this way and that's a DevOps mindset. Are you prepared for that, do you see it coming? And what's your answer to that? >> So we saw that coming about 3 and a half plus years ago. And our shift away from being a pure SI began then. And so we are an SI, but we are a service integrator rather than a systems integrator. And we began that trend in our journey, 3 plus years ago. And the reason we began that trend was what you pointed out. Today, infrastructure is delivered as a code. So not even as a service but as a code, and so imagine provisioning infrastructure and all the capabilities that ride on it, just as code. And that's where this is headed. In that model, we become provider and provisioner of services, rather than just a system. >> John: And the cost structure is completely changed because the services, Amazon has proven, and now IBM is following suit with their power platform and other things, that you can actually have the kind of compute but it's a catalog of services. So this is going to change the price competitiveness. So you know, big bids, that used to be billions of dollars, you guys can compete. I mean, am I seeing it right? >> Brian: That transition's already, that ship's sailed, so to speak, in terms of the large outsourcing deals the large, where there's apps or infrastructure, it's all moving to digital transformation consumption based commercial models. And it's really bionics that Bala mentioned a minute ago, that is our answer to the threat you described a minute ago. It's really about automating and digitizing and building intelligence into the entire, if you will, build, deliver, operate value chain of our business. >> John: Talk about the multi-vendor, multi-choice, technology-choice, as your customers and people in general on this journey of digital transformation. They have to make, they used to make technology decisions. Now they're making business logic decisions around how to reconfigure their value chains to optimize for new efficiencies and extract away inefficiencies. Blockchain is a great example, AI is another, automation is in the middle, all the cloud. So you have now business logic as the risk, technology not so much because infrastructure as code has proven that you can have server-less, you can have all kinds of coolness that can be managed in an agile way. So the business model aspect is key. How are you guys dealing with that, cause I know you're here at the IBM Think Show, their partner. I see you at the Amazon shows. We see you guys everywhere. So you're horizontally scaling. By design, is that what customers want? What is the DXC view on this? >> So our value proposition has always had partners as the key element of what we do. And so if you look at what we do, you can look at it from two perspectives. One, proprietary ways of thinking, proprietary systems are long since gone. >> And waterfall methodologies, gone, dead. >> Yes, those are all long since gone. >> If you're still doing that, note to self: you're going to be out of business. >> Exactly, so we've actually hinged a lot of what we do on our offerings, our capabilities, and so on around openness, around open source, and so forth. So that's number one. Number Two: In this world, it's no longer about just DXC or just IBM or just somebody, one person bringing everything to our clients. It's about how do you engage proactively and build co-innovation and co-services with our partners and bring that to our clients. >> I mean, IBM just announced that a deal with Google. They've got tensorflow and their deal. So you have all kinds of melting pot. Okay, let's talk about blockchain again. Go back to my favorite topic. So, if you look up that stack, you've got blockchain, you've got cryptocurrency, protocols, and what-not, mentioned securitization, you've got security tokens, you've got utility tokens. You can almost see where this is going. And then you've got on top of that, what's coming, is a mass in-migration of decentralized application developers. Okay, kind of cloud plus. You know, they know cloud, they know DevOps, infrastructure as code, but they're looking at it from a decentralization standpoint, different makeup. And you see, ICOs, initial coin offerings, I think this is an application of you know, inefficiencies around capital markets but that's, you know, put that aside for a second. But blockchain, crypto currency, and decentralized applications, how do you guys see that trend? What are you guys doing? Are you integrating it in? You mentioned token economics, you're in the banking field. Your thoughts on that? >> Bala: Sure, on the blockchain front, as I mentioned to you, there are a number of platforms that are out there. There is the R3 Corda platform. There's a platform that JPMorgan initiated that we're leveraging as well. >> John: Yeah, so they pooh-poohed Bitcoin but then they're back in the game again. (laughter) >> Bala: Yes, that's right. And then there is the Hyperledger Fabric as well. So these platforms are going to take their course of evolution and we are working across all of those platforms. Now, the more interesting thing that you mentioned is people and skills. What we've find today in the marketplace is with our clients is a dramatic shortage of skills in these areas. And so internally, what we have done at DXC is actually open our own service delivery to a vast pool of developers that you talked about earlier as being freelance, independent folks. We open our entire service delivery to them as well. And we look at that global talent pool for our own service delivery. >> Using community as a way to scale. >> Bala: Using communities, yes. And that's exactly what we're doing in our talent process. It's not just about our people, our employees, but our partners as well as what exists in the open marketplace. >> Brian, talk about the insurance area as a way to tease out other trends. Specifically, the question is What is the biggest things that people know they're walking into? What's the tail-wind that they see, that's going to give them hope? And then, What's the head-winds? What are the blockers? And what should they be aware of? What are some of the marketplace dynamics that translate into other industries? >> Brian: Well, let's start with the obvious blocker is legacy debt, right? So you talked about the risk of all that business knowledge, that domain expertise, that's all today encapsulated in existing, what you may call legacy systems, right? So that's the head-wind by far. The tail-wind is that unlike, say 15 years ago, and we were in the last sort of, dot-com boom, when it was all about the front office and customer experience, the customer is way ahead of us. So culturally, the customer is challenging industry to catch up. So that's the tail-wind in my mind. And the real opportunity is to think about it in terms of a dual agenda. So think about it in terms as progressively, simultaneously building new digital capability, whilst ultimately beginning to unbundle and tackle that legacy debt. And I think customers now are starting to see a path forward. We're in the market in both banking and insurance with digital platforms, with industry resource models, API fabrics that can go back in, modernize legacy systems. So there's a real fast time to market. >> And it changes your engagement with clients. It's not a one and done, you're sticking through the service layer. >> Brian: Oh it's a journey, but the difference, I think, between DXC and a lot of other people is that we are in the market, in production, with real assets. And you can show that journey. So it just becomes a conversation around what's your pain point? Where are you starting from? Where do you want to go? >> And you're bringing the community in to help on the delivery side, everyone wins. >> Brian: And that community is a combination of three things. That's our own employees, obviously within the industry, and within our offerings that know banking, that know insurance. It's all of the DXC people in the horizontals. Because we're bringing everything now. These platforms encapsulate infrastructure, security, service management, analytics, mobility, all of that is built into these platforms. And then, it's going out into our partner community. And then, it's going out into the open community. And we're tapping into all of those. >> John: Brian and Bala, thanks so much. 2 power CTOs here on the Cube, having a CTO conversation around how scale, cloud, AI, blockchain, new technologies are enabling new business models at a faster pace of change, with a lower cost structure, and more time to value. Again, it's all about the value creation. The killer app is money and marketplaces and community. Guys, thanks so much for sharing. I'm John Furrier here at IBM Think 2018 Cube Studios. More after this short break. (electronic music)