>>Hi, this is Cindy Mikey, vice president of industry solutions at caldera. Joining me today is chef is Molly, our solution engineer for the public sector. Today. We're going to talk about speed to insight. Why using machine learning in the public sector, specifically around fraud, waste and abuse. So topic for today, we'll discuss machine learning, why the public sector uses it to target fraud, waste, and abuse, the challenges. How do we enhance your data and analytical approaches the data landscape analytical methods and shad we'll go over reference architecture and a case study. So by definition at fraud waste and abuse per the government accountability office is broad as an attempt to obtain something about a value through unwelcomed misrepresentation waste is about squandering money or resources and abuse is about behaving improperly or unreasonably to actually obtain something of value for your personal, uh, benefit. So as we look at fraud, um, and across all industries, it's a top of mind, um, area within the public sector. >>Um, the types of fraud that we see is specifically around cyber crime, uh, looking at accounting fraud, whether it be from an individual perspective to also, uh, within organizations, looking at financial statement fraud, to also looking at bribery and corruption, as we look at fraud, it really hits us from all angles, whether it be from external perpetrators or internal perpetrators, and specifically for the research by PWC, the key focus area is we also see over half of fraud is actually through some form of internal or external perpetrators, again, key topics. So as we also look at a report recently by the association of certified fraud examiners, um, within the public sector, the us government, um, in 2017, it was identified roughly $148 billion was attributable to fraud, waste and abuse. Specifically of that 57 billion was focused on reported monetary losses and another 91 billion on areas where that opportunity or the monetary basis had not yet been measured. >>As we look at breaking those areas down again, we look at several different topics from an out payment perspective. So breaking it down within the health system, over $65 billion within social services, over $51 billion to procurement fraud to also, uh, uh, fraud, waste and abuse that's happening in the grants and the loan process to payroll fraud, and then other aspects, again, quite a few different topical areas. So as we look at those areas, what are the areas that we see additional type of focus, those are broad stroke areas. What are the actual use cases that, um, agencies are using the data landscape? What data, what analytical methods can we use to actually help curtail and prevent some of the, uh, the fraud waste and abuse. So, as we look at some of the analytical processes and analytical use great, uh, use cases in the public sector, whether it's from, uh, you know, the taxation areas to looking at, you know, social services, uh, to public safety, to also the, um, our, um, additional agency methods, we're going to focus specifically on some of the use cases around, um, you know, fraud within the tax area. >>Uh, we'll briefly look at some of the aspects of unemployment insurance fraud, uh, benefit fraud, as well as payment integrity. So fraud has its, um, uh, underpinnings in quite a few different government agencies and difficult, different analytical methods and I usage of different data. So I think one of the key elements is, you know, you can look at your, your data landscape on specific data sources that you need, but it's really about bringing together different data sources across a different variety, a different velocity. So, uh, data has different dimensions. So we'll look at on structured types of data of semi-structured data, behavioral data, as well as when we look at, um, you know, predictive models, we're typically looking at historical type information, but if we're actually trying to look at preventing fraud before it actually happens, or when a case may be in flight, which is specifically a use case that Chev is going to talk about later it's how do I look at more, that real, that streaming information? >>How do I take advantage of data, whether it be, uh, you know, uh, financial transactions we're looking at, um, asset verification, we're looking at tax records, we're looking at corporate filings. Um, and we can also look at more, uh, advanced data sources where as we're looking at, um, investigation type information. So we're maybe going out and we're looking at, uh, deep learning type models around, uh, you know, semi or that, uh, behavioral that's unstructured data, whether it be camera analysis and so forth. So for quite a different variety of data and the breadth and the opportunity really comes about when you can integrate and look at data across all different data sources. So in essence, looking at a more extensive, uh, data landscape. So specifically I want to focus on some of the methods, some of the data sources and some of the analytical techniques that we're seeing, uh, being used, um, in the government agencies, as well as opportunities to look at new methods. >>So as we're looking at, you know, from a, um, an audit planning or looking at, uh, the opportunity for the likelihood of non-compliance, um, specifically we'll see data sources where we're maybe looking at a constituents profile, we might actually be investigating the forms that they provided. We might be comparing that data, um, or leveraging internal data sources, possibly looking at net worth, comparing it against other financial data, and also comparison across other constituents groups. Some of the techniques that we use are some of the basic natural language processing, maybe we're going to do some text mining. We might be doing some probabilistic modeling, uh, where we're actually looking at, um, information within the agency to also comparing that against possibly tax forms. A lot of times it's information historically has been done on a batch perspective, both structured and semi-structured type information. And typically the data volumes can be low, but we're also seeing those data volumes on increase exponentially based upon the types of events that we're dealing with, the number of transactions. >>Um, so getting the throughput, um, and chef's going to specifically talk about that in a moment. The other aspect is, as we look at other areas of opportunity is when we're building upon, how do I actually do compliance? How do I actually look at conducting audits or potential fraud to also looking at areas of under-reported tax information? So there you might be pulling in, um, some of our other types of data sources, whether it's being property records, it could be data that's being supplied by the actual constituents or by vendors to also pulling in social media information to geographical information, to leveraging photos on techniques that we're seeing used is possibly some sentiment analysis, link analysis. Um, how do we actually blend those data sources together from a natural language processing? But I think what's important here is also the method and the looking at the data velocity, whether it be batch, whether it be near real time, again, looking at all types of data, whether it's structured semi-structured or unstructured and the key and the value behind this is, um, how do we actually look at increasing the potential revenue or the, uh, under reported revenue? >>Uh, how do we actually look at stopping fraudulent payments before they actually occur? Um, also looking at increasing the amount of, uh, the level of compliance, um, and also looking at the potential of prosecution of fraud cases. And additionally, other areas of opportunity could be looking at, um, economic planning. How do we actually perform some link analysis? How do we bring some more of those things that we saw in the data landscape on customer, or, you know, constituent interaction, bringing in social media, bringing in, uh, potentially police records, property records, um, other tax department, database information. Um, and then also looking at comparing one individual to other individuals, looking at people like a specific constituent, are there areas where we're seeing, uh, um, other aspects of a fraud potentially being occurring. Um, and also as we move forward, some of the more advanced techniques that we're seeing around deep learning is looking at computer vision, um, leveraging geospatial information, looking at social network entity analysis, uh, also looking at, um, agent-based modeling techniques, where we're looking at, uh, simulation Monte Carlo type techniques that we typically see in the financial services industry, actually applying that to fraud, waste, and abuse within the, uh, the public sector. >>Um, and again, that really lends itself to a new opportunities. And on that, I'm going to turn it over to Shev to talk about, uh, the reference architecture for, uh, doing these baskets. >>Thanks, Cindy. Um, so I'm going to walk you through an example, reference architecture for fraud detection using, uh, Cloudera underlying technology. Um, and you know, before I get into the technical details, uh, I want to talk about how this would be implemented at a much higher level. So with fraud detection, what we're trying to do is identify anomalies or novelists behavior within our data sets. Um, now in order to understand what aspects of our incoming data represents anomalous behavior, we first need to understand what normal behavior is. So in essence, once we understand normal behavior, anything that deviates from it can be thought of as an anomaly, right? So in order to understand what normal behavior is, we're going to need to be able to collect store and process a very large amount of historical data. And so then comes clutter's platform and this reference architecture that needs to before you, so, uh, let's start on the left-hand side of this reference architecture with the collect phase. >>So fraud detection will always begin with data collection. Uh, we need to collect large amounts of information from systems that could be in the cloud. It could be in the data center or even on edge devices, and this data needs to be collected so we can create our normal behavior profiles. And these normal behavioral profiles would then in turn, be used to create our predictive models for fraudulent activity. Now, uh, uh, to the data collection side, one of the main challenges that many organizations face, uh, in this phase, uh, involves using a single technology that can handle, uh, data that's coming in all different types of formats and protocols and standards with different porosities and velocities. Um, let me give you an example. Uh, we could be collecting data from a database that gets updated daily, uh, and maybe that data is being collected in Agra format. >>At the same time, we can be collecting data from an edge device that's streaming in every second, and that data may be coming in Jason or a binary format, right? So this is a data collection challenge that can be solved with clutter data flow, which is a suite of technologies built on Apache NIFA and mini five, allowing us to ingest all of this data, do a drag and drop interface. So now we're collecting all of this data, that's required to map out normal behavior. The next thing that we need to do is enrich it, transform it and distribute it to, uh, you know, downstream systems for further process. Uh, so let's, let's walk through how that would work first. Let's taking Richmond for, uh, for enrichment, think of adding additional information to your incoming data, right? Let's take, uh, financial transactions, for example, uh, because Cindy mentioned it earlier, right? >>You can store known locations of an individual in an operational database, uh, with Cloudera that would be HBase. And as an individual makes a new transaction, their geo location that's in that transaction data, it can be enriched with previously known locations of that very same individual and all of that enriched data. It can be later used downstream for predictive analysis, predictable. So the data has been enrich. Uh, now it needs to be transformed. We want the data that's coming in, uh, you know, Avro and Jason and binary and whatever other format to be transformed into a single common format. So it can be used downstream for stream processing. Uh, again, this is going to be done through clutter and data flow, which is backed by NIFA, right? So the transformed semantic data is then going to be stimulated to Kafka and coffin. It's going to serve as that central repository of syndicated services or a buffer zone, right? >>So cough is, you know, pretty much provides you with, uh, extremely fast resilient and fault tolerance storage. And it's also going to give you the consumer APIs that you need that are going to enable a wide variety of applications to leverage that enriched and transformed data within your buffer zone. Uh, I'll add that, you know, 17, so you can store that data, uh, in a distributed file system, give you that historical context that you're going to need later on for machine learning, right? So the next step in the architecture is to leverage a cluttered SQL string builder, which enables us to write, uh, streaming sequel jobs on top of Apache Flink. So we can, uh, filter, analyze and, uh, understand the data that's in the Kafka buffer zone in real time. Uh I'll you know, I'll also add like, you know, if you have time series data, or if you need a lab type of cubing, you can leverage kudu, uh, while EDA or exploratory data analysis and visualization, uh, can all be enabled through clever visual patient technology. >>All right, so we've filtered, we've analyzed and we've explored our incoming data. We can now proceed to train our machine learning models, uh, which will detect anomalous behavior in our historically collected data set, uh, to do this, we can use a combination of supervised unsupervised, uh, even deep learning techniques with neural networks and these models can be tested on new incoming streaming data. And once we've gone ahead and obtain the accuracy of the performance, the scores that we want, we can then take these models and deploy them into production. And once the models are productionalized or operationalized, they can be leveraged within our streaming pipeline. So as new data is ingested in real-time knife, I can query these models to detect if the activity is anomalous or fraudulent. And if it is, they can alert downstream users and systems, right? So this in essence is how fraudulent activity detection works. >>Uh, and this entire pipeline is powered by clutter's technology, right? And so, uh, the IRS is one of, uh, clutters customers. That's leveraging our platform today and implementing, uh, a very similar architecture, uh, to detect fraud, waste, and abuse across a very large set of, uh, historical facts, data. Um, and one of the neat things with the IRS is that they've actually, uh, recently leveraged the partnership between Cloudera and Nvidia to accelerate their Spark-based analytics and their machine learning. Uh, and the results have been nothing short of amazing, right? And in fact, we have a quote here from Joe and salty who's, uh, you know, the technical branch chief for the research analytics and statistics division group within the IRS with zero changes to our fraud detection workflow, we're able to obtain eight times to performance simply by adding GPS to our mainstream big data servers. This improvement translates to half the cost of ownership for the same workloads, right? So embedding GPU's into the reference architecture I covered earlier has enabled the IRS to improve their time to insights by as much as eight X while simultaneously reducing their underlying infrastructure costs by half, uh, Cindy back to you >>Chef. Thank you. Um, and I hope that you found, uh, some of the, the analysis, the information that Sheva and I have provided, um, to give you some insights on how cloud era is actually helping, uh, with the fraud waste and abuse challenges within the, uh, the public sector, um, specifically looking at any and all types of data, how the clutter a platform is bringing together and analyzing information, whether it be you're structured you're semi-structured to unstructured data, both in a fast or in a real time perspective, looking at anomalies, being able to do some of those on detection methods, uh, looking at neural network analysis, time series information. So next steps we'd love to have an additional conversation with you. You can also find on some additional information around, uh, how quad areas working in the federal government by going to cloudera.com solutions slash public sector. And we welcome scheduling a meeting with you again, thank you for joining Chevy and I today, we greatly appreciate your time and look forward to future >>Conversation..

>>Hi, this is Cindy Mikey, vice president of industry solutions at caldera. Joining me today is chef is Molly, our solution engineer for the public sector. Today. We're going to talk about speed to insight. Why using machine learning in the public sector, specifically around fraud, waste and abuse. So topic for today, we'll discuss machine learning, why the public sector uses it to target fraud, waste, and abuse, the challenges. How do we enhance your data and analytical approaches the data landscape analytical methods and shad we'll go over reference architecture and a case study. So by definition, fraud, waste and abuse per the government accountability office is fraud. Isn't an attempt to obtain something about value through unwelcome misrepresentation waste is about squandering money or resources and abuse is about behaving improperly or unreasonably to actually obtain something of value for your personal benefit. So as we look at fraud, um, and across all industries, it's a top of mind, um, area within the public sector. >>Um, the types of fraud that we see is specifically around cyber crime, uh, looking at accounting fraud, whether it be from an individual perspective to also, uh, within organizations, looking at financial statement fraud, to also looking at bribery and corruption, as we look at fraud, it really hits us from all angles, whether it be from external perpetrators or internal perpetrators, and specifically from the research by PWC, the key focus area is we also see over half of fraud is actually through some form of internal or external, uh, perpetrators again, key topics. So as we also look at a report recently by the association of certified fraud examiners, um, within the public sector, the us government, um, in 2017, it was identified roughly $148 billion was attributable to fraud, waste and abuse. Specifically about 57 billion was focused on reported monetary losses and another 91 billion on areas where that opportunity or the monetary basis had not yet been measured. >>As we look at breaking those areas down again, we look at several different topics from permit out payment perspective. So breaking it down within the health system, over $65 billion within social services, over $51 billion to procurement fraud to also, um, uh, fraud, waste and abuse that's happening in the grants and the loan process to payroll fraud, and then other aspects, again, quite a few different topical areas. So as we look at those areas, what are the areas that we see additional type of focus, there's a broad stroke areas. What are the actual use cases that our agencies are using the data landscape? What data, what analytical methods can we use to actually help curtail and prevent some of the, uh, the fraud waste and abuse. So, as we look at some of the analytical processes and analytical use crate, uh, use cases in the public sector, whether it's from, uh, you know, the taxation areas to looking at, you know, social services, uh, to public safety, to also the, um, our, um, uh, additional agency methods, we're gonna use focused specifically on some of the use cases around, um, you know, fraud within the tax area. >>Uh, we'll briefly look at some of the aspects of, um, unemployment insurance fraud, uh, benefit fraud, as well as payment and integrity. So fraud has it it's, um, uh, underpinnings inquiry, like you different on government agencies and difficult, different analytical methods, and I usage of different data. So I think one of the key elements is, you know, you can look at your, your data landscape on specific data sources that you need, but it's really about bringing together different data sources across a different variety, a different velocity. So, uh, data has different dimensions. So we'll look at structured types of data of semi-structured data, behavioral data, as well as when we look at, um, you know, predictive models. We're typically looking at historical type information, but if we're actually trying to look at preventing fraud before it actually happens, or when a case may be in flight, which is specifically a use case that shad is going to talk about later is how do I look at more of that? >>Real-time that streaming information? How do I take advantage of data, whether it be, uh, you know, uh, financial transactions we're looking at, um, asset verification, we're looking at tax records, we're looking at corporate filings. Um, and we can also look at more, uh, advanced data sources where as we're looking at, um, investigation type information. So we're maybe going out and we're looking at, uh, deep learning type models around, uh, you know, semi or that, uh, behavioral, uh, that's unstructured data, whether it be camera analysis and so forth. So for quite a different variety of data and the, the breadth and the opportunity really comes about when you can integrate and look at data across all different data sources. So in a looking at a more extensive, uh, data landscape. So specifically I want to focus on some of the methods, some of the data sources and some of the analytical techniques that we're seeing, uh, being used, um, in the government agencies, as well as opportunities, uh, to look at new methods. >>So as we're looking at, you know, from a, um, an audit planning or looking at, uh, the opportunity for the likelihood of non-compliance, um, specifically we'll see data sources where we're maybe looking at a constituents profile, we might actually be investigating the forms that they've provided. We might be comparing that data, um, or leveraging internal data sources, possibly looking at net worth, comparing it against other financial data, and also comparison across other constituents groups. Some of the techniques that we use are some of the basic natural language processing, maybe we're going to do some text mining. We might be doing some probabilistic modeling, uh, where we're actually looking at, um, information within the agency to also comparing that against possibly tax forms. A lot of times it's information historically has been done on a batch perspective, both structured and semi-structured type information. And typically the data volumes can be low, but we're also seeing those data volumes on increase exponentially based upon the types of events that we're dealing with, the number of transactions. >>Um, so getting the throughput, um, and chef's going to specifically talk about that in a moment. The other aspect is, as we look at other areas of opportunity is when we're building upon, how do I actually do compliance? How do I actually look at conducting audits, uh, or potential fraud to also looking at areas of under-reported tax information? So there you might be pulling in some of our other types of data sources, whether it's being property records, it could be data that's being supplied by the actual constituents or by vendors to also pulling in social media information to geographical information, to leveraging photos on techniques that we're seeing used is possibly some sentiment analysis, link analysis. Um, how do we actually blend those data sources together from a natural language processing? But I think what's important here is also the method and the looking at the data velocity, whether it be batch, whether it be near real time, again, looking at all types of data, whether it's structured semi-structured or unstructured and the key and the value behind this is, um, how do we actually look at increasing the potential revenue or the, um, under reported revenue? >>Uh, how do we actually look at stopping fraudulent payments before they actually occur? Um, also looking at increasing the amount of, uh, the level of compliance, um, and also looking at the potential of prosecution of fraud cases. And additionally, other areas of opportunity could be looking at, um, economic planning. How do we actually perform some link analysis? How do we bring some more of those things that we saw in the data landscape on customer, or, you know, constituent interaction, bringing in social media, bringing in, uh, potentially police records, property records, um, other tax department, database information. Um, and then also looking at comparing one individual to other individuals, looking at people like a specific, like a constituent, are there areas where we're seeing, uh, >>Um, other >>Aspects of, of fraud potentially being occurring. Um, and also as we move forward, some of the more advanced techniques that we're seeing around deep learning is looking at computer vision, um, leveraging geospatial information, looking at social network entity analysis, uh, also looking at, uh, agent-based modeling techniques, where we're looking at simulation Monte Carlo type techniques that we typically see in the financial services industry, actually applying that to fraud, waste, and abuse within the, uh, the public sector. Um, and again, that really, uh, lends itself to a new opportunities. And on that, I'm going to turn it over to chef to talk about, uh, the reference architecture for, uh, doing these buckets. >>Thanks, Cindy. Um, so I'm gonna walk you through an example, reference architecture for fraud detection using, uh, Cloudera's underlying technology. Um, and you know, before I get into the technical details, uh, I want to talk about how this would be implemented at a much higher level. So with fraud detection, what we're trying to do is identify anomalies or novelists behavior within our datasets. Um, now in order to understand what aspects of our incoming data represents anomalous behavior, we first need to understand what normal behavior is. So in essence, once we understand normal behavior, anything that deviates from it can be thought of as an anomaly, right? So in order to understand what normal behavior is, we're going to need to be able to collect store and process a very large amount of historical data. And so incomes, clutters platform, and this reference architecture that needs to be for you. >>So, uh, let's start on the left-hand side of this reference architecture with the collect phase. So fraud detection will always begin with data collection. We need to collect large amounts of information from systems that could be in the cloud. It could be in the data center or even on edge devices, and this data needs to be collected so we can create our normal behavior profiles. And these normal behavioral profiles would then in turn, be used to create our predictive models for fraudulent activity. Now, uh, thinking, uh, to the data collection side, one of the main challenges that many organizations face, uh, in this phase, uh, involves using a single technology that can handle, uh, data that's coming in all different types of formats and protocols and standards with different velocities and velocities. Um, let me give you an example. Uh, we could be collecting data from a database that gets updated daily, uh, and maybe that data is being collected in Agra format. >>At the same time, we can be collecting data from an edge device that's streaming in every second, and that data may be coming in Jason or a binary format, right? So this is a data collection challenge that can be solved with cluttered data flow, which is a suite of technologies built on a patch NIFA in mini five, allowing us to ingest all of this data, do a drag and drop interface. So now we're collecting all of this data, that's required to map out normal behavior. The next thing that we need to do is enrich it, transform it and distribute it to, uh, you know, downstream systems for further process. Uh, so let's, let's walk through how that would work first. Let's taking Richmond for, uh, for enrichment, think of adding additional information to your incoming data, right? Let's take, uh, financial transactions, for example, uh, because Cindy mentioned it earlier, right? >>You can store known locations of an individual in an operational database, uh, with Cloudera that would be HBase. And as an individual makes a new transaction, their geolocation that's in that transaction data can be enriched with previously known locations of that very same individual. And all of that enriched data can be later used downstream for predictive analysis, predictable. So the data has been enrich. Uh, now it needs to be transformed. We want the data that's coming in, uh, you know, Avro and Jason and binary and whatever other format to be transformed into a single common format. So it can be used downstream for stream processing. Uh, again, this is going to be done through clutter and data flow, which is backed by NIFA, right? So the transformed semantic data is then going to be stricted to Kafka and coffin. It's going to serve as that central repository of syndicated services or a buffer zone, right? >>So coffee is going to pretty much provide you with, uh, extremely fast resilient and fault tolerance storage. And it's also gonna give you the consumer APIs that you need that are going to enable a wide variety of applications to leverage that enriched and transformed data within your buffer zone, uh, allowed that, you know, 17. So you can store that data in a distributed file system, give you that historical context that you're going to need later on for machine learning, right? So the next step in the architecture is to leverage a cluttered SQL stream builder, which enables us to write, uh, streaming SQL jobs on top of Apache Flink. So we can, uh, filter, analyze and, uh, understand the data that's in the Kafka buffer in real time. Uh I'll you know, I'll also add like, you know, if you have time series data, or if you need a lab type of cubing, you can leverage kudu, uh, while EDA or, you know, exploratory data analysis and visualization, uh, can all be enabled through clever visualization technology. >>All right, so we've filtered, we've analyzed and we've explored our incoming data. We can now proceed to train our machine learning models, uh, which will detect anomalous behavior in our historically collected data set, uh, to do this, we can use a combination of supervised unsupervised, uh, even deep learning techniques with neural networks. And these models can be tested on new incoming streaming data. And once we've gone ahead and obtain the accuracy of the performance, the scores that we want, we can then take these models and deploy them into production. And once the models are productionalized or operationalized, they can be leveraged within our streaming pipeline. So as new data is ingested in real-time knife, I can query these models to detect if the activity is anomalous or fraudulent. And if it is, they can alert downstream users and systems, right? So this in essence is how fraudulent activity detection works. >>Uh, and this entire pipeline is powered by clutters technology, right? And so, uh, the IRS is one of, uh, clutter's customers. That's leveraging our platform today and implementing, uh, a very similar architecture, uh, to detect fraud, waste, and abuse across a very large set of historical facts, data. Um, and one of the neat things with the IRS is that they've actually recently leveraged the partnership between Cloudera and Nvidia to accelerate their spark based analytics and their machine learning, uh, and the results have been nothing short of amazing, right? And in fact, we have a quote here from Joe and salty who's, uh, you know, the technical branch chief for the research analytics and statistics division group within the IRS with zero changes to our fraud detection workflow, we're able to obtain eight times to performance simply by adding GPS to our mainstream big data servers. This improvement translates to half the cost of ownership for the same workloads, right? So embedding GPU's into the reference architecture I covered earlier has enabled the IRS to improve their time to insights by as much as eight X while simultaneously reducing their underlying infrastructure costs by half, uh, Cindy back to you >>Chef. Thank you. Um, and I hope that you found, uh, some of the, the analysis, the information that Sheva and I have provided, um, to give you some insights on how cloud era is actually helping, uh, with the fraud waste and abuse challenges within the, uh, the public sector, um, specifically looking at any and all types of data, how the clutter platform is bringing together and analyzing information, whether it be you're structured you're semi-structured to unstructured data, both in a fast or in a real-time perspective, looking at anomalies, being able to do some of those on detection, uh, looking at neural network analysis, time series information. So next steps we'd love to have additional conversation with you. You can also find on some additional information around, I have caught areas working in the, the federal government by going to cloudera.com solutions slash public sector. And we welcome scheduling a meeting with you again, thank you for joining us Sheva and I today. We greatly appreciate your time and look forward to future progress. >>Good day, everyone. Thank you for joining me. I'm Sydney. Mike joined by Rick Taylor of Cloudera. Uh, we're here to talk about predictive maintenance for the public sector and how to increase assets, service, reliability on today's agenda. We'll talk specifically around how to optimize your equipment maintenance, how to reduce costs, asset failure with data and analytics. We'll go into a little more depth on, um, what type of data, the analytical methods that we're typically seeing used, um, the associated, uh, Brooke, we'll go over a case study as well as a reference architecture. So by basic definition, uh, predictive maintenance is about determining when an asset should be maintained and what specific maintenance activities need to be performed either based upon an assets of actual condition or state. It's also about predicting and preventing failures and performing maintenance on your time on your schedule to avoid costly unplanned downtime. >>McKinsey has looked at analyzing predictive maintenance costs across multiple industries and has identified that there's the opportunity to reduce overall predictive maintenance costs by roughly 50% with different types of analytical methods. So let's look at those three types of models. First, we've got our traditional type of method for maintenance, and that's really about our corrective maintenance, and that's when we're performing maintenance on an asset, um, after the equipment fails. But the challenges with that is we end up with unplanned. We end up with disruptions in our schedules, um, as well as reduced quality, um, around the performance of the asset. And then we started looking at preventive maintenance and preventative maintenance is really when we're performing maintenance on a set schedule. Um, the challenges with that is we're typically doing it regardless of the actual condition of the asset, um, which has resulted in unnecessary downtime and expense. Um, and specifically we're really now focused on pre uh, condition-based maintenance, which is looking at leveraging predictive maintenance techniques based upon actual conditions and real time events and processes. Um, within that we've seen organizations, um, and again, source from McKenzie have a 50% reduction in downtime, as well as an overall 40% reduction in maintenance costs. Again, this is really looking at things across multiple industries, but let's look at it in the context of the public sector and based upon some activity by the department of energy, um, several years ago, >>Um, they've really >>Looked at what does predictive maintenance mean to the public sector? What is the benefit, uh, looking at increasing return on investment of assets, reducing, uh, you know, reduction in downtime, um, as well as overall maintenance costs. So corrective or reactive based maintenance is really about performing once there's been a failure. Um, and then the movement towards, uh, preventative, which is based upon a set schedule or looking at predictive where we're monitoring real-time conditions. Um, and most importantly is now actually leveraging IOT and data and analytics to further reduce those overall downtimes. And there's a research report by the, uh, department of energy that goes into more specifics, um, on the opportunity within the public sector. So, Rick, let's talk a little bit about what are some of the challenges, uh, regarding data, uh, regarding predictive maintenance. >>Some of the challenges include having data silos, historically our government organizations and organizations in the commercial space as well, have multiple data silos. They've spun up over time. There are multiple business units and note, there's no single view of assets. And oftentimes there's redundant information stored in, in these silos of information. Uh, couple that with huge increases in data volume data growing exponentially, along with new types of data that we can ingest there's social media, there's semi and unstructured data sources and the real time data that we can now collect from the internet of things. And so the challenge is to collect all these assets together and begin to extract intelligence from them and insights and, and that in turn then fuels, uh, machine learning and, um, and, and what we call artificial intelligence, which enables predictive maintenance. Next slide. So >>Let's look specifically at, you know, the, the types of use cases and I'm going to Rick and I are going to focus on those use cases, where do we see predictive maintenance coming into the procurement facility, supply chain, operations and logistics. Um, we've got various level of maturity. So, you know, we're talking about predictive maintenance. We're also talking about, uh, using, uh, information, whether it be on a, um, a connected asset or a vehicle doing monitoring, uh, to also leveraging predictive maintenance on how do we bring about, uh, looking at data from connected warehouses facilities and buildings all bring on an opportunity to both increase the quality and effectiveness of the missions within the agencies to also looking at re uh, looking at cost efficiency, as well as looking at risk and safety and the types of data, um, you know, that Rick mentioned around, you know, the new types of information, some of those data elements that we typically have seen is looking at failure history. >>So when has that an asset or a machine or a component within a machine failed in the past? Uh, we've also looking at bringing together a maintenance history, looking at a specific machine. Are we getting error codes off of a machine or assets, uh, looking at when we've replaced certain components to looking at, um, how are we actually leveraging the assets? What were the operating conditions, uh, um, pulling off data from a sensor on that asset? Um, also looking at the, um, the features of an asset, whether it's, you know, engine size it's make and model, um, where's the asset located on to also looking at who's operated the asset, uh, you know, whether it be their certifications, what's their experience, um, how are they leveraging the assets and then also bringing in together, um, some of the, the pattern analysis that we've seen. So what are the operating limits? Um, are we getting service reliability? Are we getting a product recall information from the actual manufacturer? So, Rick, I know the data landscape has really changed. Let's, let's go over looking at some of those components. Sure. >>So this slide depicts sort of the, some of the inputs that inform a predictive maintenance program. So, as we've talked a little bit about the silos of information, the ERP system of record, perhaps the spares and the service history. So we want, what we want to do is combine that information with sensor data, whether it's a facility and equipment sensors, um, uh, or temperature and humidity, for example, all this stuff is then combined together, uh, and then use to develop machine learning models that better inform, uh, predictive maintenance, because we'll do need to keep, uh, to take into account the environmental factors that may cause additional wear and tear on the asset that we're monitoring. So here's some examples of private sector, uh, maintenance use cases that also have broad applicability across the government. For example, one of the busiest airports in Europe is running cloud era on Azure to capture secure and correlate sensor data collected from equipment within the airport, the people moving equipment more specifically, the escalators, the elevators, and the baggage carousels. >>The objective here is to prevent breakdowns and improve airport efficiency and passenger safety. Another example is a container shipping port. In this case, we use IOT data and machine learning, help customers recognize how their cargo handling equipment is performing in different weather conditions to understand how usage relates to failure rates and to detect anomalies and transport systems. These all improve for another example is Navistar Navistar, leading manufacturer of commercial trucks, buses, and military vehicles. Typically vehicle maintenance, as Cindy mentioned, is based on miles traveled or based on a schedule or a time since the last service. But these are only two of the thousands of data points that can signal the need for maintenance. And as it turns out, unscheduled maintenance and vehicle breakdowns account for a large share of the total cost for vehicle owner. So to help fleet owners move from a reactive approach to a more predictive model, Navistar built an IOT enabled remote diagnostics platform called on command. >>The platform brings in over 70 sensor data feeds for more than 375,000 connected vehicles. These include engine performance, trucks, speed, acceleration, cooling temperature, and break where this data is then correlated with other Navistar and third-party data sources, including weather geo location, vehicle usage, traffic warranty, and parts inventory information. So the platform then uses machine learning and advanced analytics to automatically detect problems early and predict maintenance requirements. So how does the fleet operator use this information? They can monitor truck health and performance from smartphones or tablets and prioritize needed repairs. Also, they can identify that the nearest service location that has the relevant parts, the train technicians and the available service space. So sort of wrapping up the, the benefits Navistar's helped fleet owners reduce maintenance by more than 30%. The same platform is also used to help school buses run safely. And on time, for example, one school district with 110 buses that travel over a million miles annually reduce the number of PTOs needed year over year, thanks to predictive insights delivered by this platform. >>So I'd like to take a moment and walk through the data. Life cycle is depicted in this diagram. So data ingest from the edge may include feeds from the factory floor or things like connected vehicles, whether they're trucks, aircraft, heavy equipment, cargo vessels, et cetera. Next, the data lands on a secure and governed data platform. Whereas combined with data from existing systems of record to provide additional insights, and this platform supports multiple analytic functions working together on the same data while maintaining strict security governance and control measures once processed the data is used to train machine learning models, which are then deployed into production, monitored, and retrained as needed to maintain accuracy. The process data is also typically placed in a data warehouse and use to support business intelligence, analytics, and dashboards. And in fact, this data lifecycle is representative of one of our government customers doing condition-based maintenance across a variety of aircraft. >>And the benefits they've discovered include less unscheduled maintenance and a reduction in mean man hours to repair increased maintenance efficiencies, improved aircraft availability, and the ability to avoid cascading component failures, which typically cost more in repair cost and downtime. Also, they're able to better forecast the requirements for replacement parts and consumables and last, and certainly very importantly, this leads to enhanced safety. This chart overlays the secure open source Cloudera platform used in support of the data life cycle. We've been discussing Cloudera data flow, the data ingest data movement and real time streaming data query capabilities. So data flow gives us the capability to bring data in from the asset of interest from the internet of things. While the data platform provides a secure governed data lake and visibility across the full machine learning life cycle eliminates silos and streamlines workflows across teams. The platform includes an integrated suite of secure analytic applications. And two that we're specifically calling out here are Cloudera machine learning, which supports the collaborative data science and machine learning environment, which facilitates machine learning and AI and the cloud era data warehouse, which supports the analytics and business intelligence, including those dashboards for leadership Cindy, over to you, Rick, >>Thank you. And I hope that, uh, Rick and I provided you some insights on how predictive maintenance condition-based maintenance is being used and can be used within your respective agency, bringing together, um, data sources that maybe you're having challenges with today. Uh, bringing that, uh, more real-time information in from a streaming perspective, blending that industrial IOT, as well as historical information together to help actually, uh, optimize maintenance and reduce costs within the, uh, each of your agencies, uh, to learn a little bit more about Cloudera, um, and our, what we're doing from a predictive maintenance please, uh, business@cloudera.com solutions slash public sector. And we look forward to scheduling a meeting with you, and on that, we appreciate your time today and thank you very much.

>>Hi, this is Cindy Mikey, vice president of industry solutions at Cloudera. Joining me today is chef is Molly, our solution engineer for the public sector. Today. We're going to talk about speed to insight. Why using machine learning in the public sector, specifically around fraud, waste and abuse. So topic for today, we'll discuss machine learning, why the public sector uses it to target fraud, waste, and abuse, the challenges. How do we enhance your data and analytical approaches the data landscape analytical methods and Shev we'll go over reference architecture and a case study. So by definition, fraud, waste and abuse per the government accountability office is fraud is an attempt to obtain something about a value through unwelcomed. Misrepresentation waste is about squandering money or resources and abuse is about behaving improperly or unreasonably to actually obtain something of value for your personal benefit. So as we look at fraud and across all industries, it's a top of mind, um, area within the public sector. >>Um, the types of fraud that we see is specifically around cyber crime, uh, looking at accounting fraud, whether it be from an individual perspective to also, uh, within organizations, looking at financial statement fraud, to also looking at bribery and corruption, as we look at fraud, it really hits us from all angles, whether it be from external perpetrators or internal perpetrators, and specifically from the research by PWC, the key focus area is we also see over half of fraud is actually through some form of internal or external are perpetrators again, key topics. So as we also look at a report recently by the association of certified fraud examiners, um, within the public sector, the us government, um, in 2017, it was identified roughly $148 billion was attributable to fraud, waste and abuse. Specifically of that 57 billion was focused on reported monetary losses and another 91 billion on areas where that opportunity or the monetary basis had not yet been measured. >>As we look at breaking those areas down again, we look at several different topics from an out payment perspective. So breaking it down within the health system, over $65 billion within social services, over $51 billion to procurement fraud to also, um, uh, fraud, waste and abuse that's happening in the grants and the loan process to payroll fraud, and then other aspects, again, quite a few different topical areas. So as we look at those areas, what are the areas that we see additional type of focus, there's broad stroke areas? What are the actual use cases that our agencies are using the data landscape? What data, what analytical methods can we use to actually help curtail and prevent some of the, uh, the fraud waste and abuse. So, as we look at some of the analytical processes and analytical use crate, uh, use cases in the public sector, whether it's from, uh, you know, the taxation areas to looking at social services, uh, to public safety, to also the, um, our, um, uh, additional agency methods, we're going to focus specifically on some of the use cases around, um, you know, fraud within the tax area. >>Uh, we'll briefly look at some of the aspects of unemployment insurance fraud, uh, benefit fraud, as well as payment and integrity. So fraud has its, um, uh, underpinnings in quite a few different on government agencies and difficult, different analytical methods and I usage of different data. So I think one of the key elements is, you know, you can look at your, your data landscape on specific data sources that you need, but it's really about bringing together different data sources across a different variety, a different velocity. So, uh, data has different dimensions. So we'll look at on structured types of data of semi-structured data, behavioral data, as well as when we look at, um, you know, predictive models, we're typically looking at historical type information, but if we're actually trying to lock at preventing fraud before it actually happens, or when a case may be in flight, which is specifically a use case, that shadow is going to talk about later it's how do I look at more of that? >>Real-time that streaming information? How do I take advantage of data, whether it be, uh, you know, uh, financial transactions we're looking at, um, asset verification, we're looking at tax records, we're looking at corporate filings. Um, and we can also look at more, uh, advanced data sources where as we're looking at, um, investigation type information. So we're maybe going out and we're looking at, uh, deep learning type models around, uh, you know, semi or that behavioral, uh, that's unstructured data, whether it be camera analysis and so forth. So quite a different variety of data and the, the breadth, um, and the opportunity really comes about when you can integrate and look at data across all different data sources. So in a sense, looking at a more extensive on data landscape. So specifically I want to focus on some of the methods, some of the data sources and some of the analytical techniques that we're seeing, uh, being used, um, in the government agencies, as well as opportunities, uh, to look at new methods. >>So as we're looking at, you know, from a, um, an audit planning or looking at, uh, the opportunity for the likelihood of non-compliance, um, specifically we'll see data sources where we're maybe looking at a constituents profile, we might actually be, um, investigating the forms that they've provided. We might be comparing that data, um, or leveraging internal data sources, possibly looking at net worth, comparing it against other financial data, and also comparison across other constituents groups. Some of the techniques that we use are some of the basic natural language processing, maybe we're going to do some text mining. We might be doing some probabilistic modeling, uh, where we're actually looking at, um, information within the agency to also comparing that against possibly tax forms. A lot of times it's information historically has been done on a batch perspective, both structured and semi-structured type information. And typically the data volumes can be low, but we're also seeing those data volumes increase exponentially based upon the types of events that we're dealing with, the number of transactions. >>Um, so getting the throughput, um, and chef's going to specifically talk about that in a moment. The other aspect is, as we look at other areas of opportunity is when we're building upon, how do I actually do compliance? How do I actually look at conducting audits, uh, or potential fraud to also looking at areas of under reported tax information? So there you might be pulling in some of our other types of data sources, whether it's being property records, it could be data that's being supplied by the actual constituents or by vendors to also pulling in social media information to geographical information, to leveraging photos on techniques that we're seeing used is possibly some sentiment analysis, link analysis. Um, how do we actually blend those data sources together from a natural language processing? But I think what's important here is also the method and the looking at the data velocity, whether it be batch, whether it be near real time, again, looking at all types of data, whether it's structured semi-structured or unstructured and the key and the value behind this is, um, how do we actually look at increasing the potential revenue or the, um, under reported revenue? >>Uh, how do we actually look at stopping fraudulent payments before they actually occur? Um, also looking at increasing the amount of, uh, the level of compliance, um, and also looking at the potential of prosecution of fraud cases. And additionally, other areas of opportunity could be looking at, um, economic planning. How do we actually perform some link analysis? How do we bring some more of those things that we saw in the data landscape on customer, or, you know, constituent interaction, bringing in social media, bringing in, uh, potentially police records, property records, um, other tax department, database information. Um, and then also looking at comparing one individual to other individuals, looking at people like a specific, like, uh, constituent, are there areas where we're seeing, uh, um, other aspects of, of fraud potentially being, uh, occurring. Um, and also as we move forward, some of the more advanced techniques that we're seeing around deep learning is looking at computer vision, um, leveraging geospatial information, looking at social network entity analysis, uh, also looking at, um, agent-based modeling techniques, where we're looking at simulation, Monte Carlo type techniques that we typically see in the financial services industry, actually applying that to fraud, waste, and abuse within the, the public sector. >>Um, and again, that really, uh, lends itself to a new opportunities. And on that, I'm going to turn it over to Chevy to talk about, uh, the reference architecture for doing these buckets. >>Sure. Yeah. Thanks, Cindy. Um, so I'm going to walk you through an example, reference architecture for fraud detection, using Cloudera as underlying technology. Um, and you know, before I get into the technical details, uh, I want to talk about how this would be implemented at a much higher level. So with fraud detection, what we're trying to do is identify anomalies or anomalous behavior within our datasets. Um, now in order to understand what aspects of our incoming data represents anomalous behavior, we first need to understand what normal behavior is. So in essence, once we understand normal behavior, anything that deviates from it can be thought of as an anomaly, right? So in order to understand what normal behavior is, we're going to need to be able to collect store and process a very large amount of historical data. And so incomes, clutters platform, and this reference architecture that needs to be for you. >>So, uh, let's start on the left-hand side of this reference architecture with the collect phase. So fraud detection will always begin with data collection. Uh, we need to collect large amounts of information from systems that could be in the cloud. It could be in the data center or even on edge devices, and this data needs to be collected so we can create from normal behavior profiles and these normal behavioral profiles would then in turn, be used to create our predictive models for fraudulent activity. Now, uh, uh, to the data collection side, one of the main challenges that many organizations face, uh, in this phase, uh, involves using a single technology that can handle, uh, data that's coming in all different types of formats and protocols and standards with different velocities and velocities. Um, let me give you an example. Uh, we could be collecting data from a database that gets updated daily, uh, and maybe that data is being collected in Agra format. >>At the same time, we can be collecting data from an edge device that's streaming in every second, and that data may be coming in Jace on or a binary format, right? So this is a data collection challenge that can be solved with cluttered data flow, which is a suite of technologies built on Apache NIFA and mini five, allowing us to ingest all of this data, do a drag and drop interface. So now we're collecting all of this data, that's required to map out normal behavior. The next thing that we need to do is enrich it, transform it and distribute it to know downstream systems for further process. Uh, so let's, let's walk through how that would work first. Let's taking Richmond for, uh, for enrichment, think of adding additional information to your incoming data, right? Let's take, uh, financial transactions, for example, uh, because Cindy mentioned it earlier, right? >>You can store known locations of an individual in an operational database, uh, with Cloudera that would be HBase. And as an individual makes a new transaction, their geo location that's in that transaction data, it can be enriched with previously known locations of that very same individual and all of that enriched data. It can be later used downstream for predictive analysis, predictable. So the data has been enrich. Uh, now it needs to be transformed. We want the data that's coming in, uh, you know, Avro and Jason and binary and whatever other format to be transformed into a single common format. So it can be used downstream for stream processing. Uh, again, this is going to be done through clutter and data flow, which is backed by NIFA, right? So the transformed semantic data is then going to be stimulated to Kafka and coffin is going to serve as that central repository of syndicated services or a buffer zone, right? >>So cough is, you know, pretty much provides you with, uh, extremely fast resilient and fault tolerance storage. And it's also going to give you the consumer API APIs that you need that are going to enable a wide variety of applications to leverage that enriched and transform data within your buffer zone. Uh, I'll add that, you know, 17, so you can store that data, uh, in a distributed file system, give you that historical context that you're going to need later on from machine learning, right? So the next step in the architecture is to leverage, uh, clutter SQL stream builder, which enables us to write, uh, streaming sequel jobs on top of Apache Flink. So we can, uh, filter, analyze and, uh, understand the data that's in the Kafka buffer zone in real-time. Uh, I'll, you know, I'll also add like, you know, if you have time series data, or if you need a lab type of cubing, you can leverage Q2, uh, while EDA or, you know, exploratory data analysis and visualization, uh, can all be enabled through clever visualization technology. >>All right, so we've filtered, we've analyzed, and we've our incoming data. We can now proceed to train our machine learning models, uh, which will detect anomalous behavior in our historically collected data set, uh, to do this, we can use a combination of supervised unsupervised, even deep learning techniques with neural networks. Uh, and these models can be tested on new incoming streaming data. And once we've gone ahead and obtain the accuracy of the performance, the X one, uh, scores that we want, we can then take these models and deploy them into production. And once the models are productionalized or operationalized, they can be leveraged within our streaming pipeline. So as new data is ingested in real time knife, I can query these models to detect if the activity is anomalous or fraudulent. And if it is, they can alert downstream users and systems, right? So this in essence is how fraudulent activity detection works. Uh, and this entire pipeline is powered by clutters technology. Uh, Cindy, next slide please. >>Right. And so, uh, the IRS is one of, uh, clutter as customers. That's leveraging our platform today and implementing a very similar architecture, uh, to detect fraud, waste, and abuse across a very large set of, uh, historical facts, data. Um, and one of the neat things with the IRS is that they've actually recently leveraged the partnership between Cloudera and Nvidia to accelerate their Spark-based analytics and their machine learning. Uh, and the results have been nothing short of amazing, right? And in fact, we have a quote here from Joe and salty who's, uh, you know, the technical branch chief for the research analytics and statistics division group within the IRS with zero changes to our fraud detection workflow, we're able to obtain eight times to performance simply by adding GPS to our mainstream big data servers. This improvement translates to half the cost of ownership for the same workloads, right? So embedding GPU's into the reference architecture I covered earlier has enabled the IRS to improve their time to insights by as much as eight X while simultaneously reducing their underlying infrastructure costs by half, uh, Cindy back to you >>Chef. Thank you. Um, and I hope that you found, uh, some of the, the analysis, the information that Sheva and I have provided, uh, to give you some insights on how cloud era is actually helping, uh, with the fraud waste and abuse challenges within the, uh, the public sector, um, specifically looking at any and all types of data, how the clutter a platform is bringing together and analyzing information, whether it be you're structured you're semi-structured to unstructured data, both in a fast or in a real-time perspective, looking at anomalies, being able to do some of those on detection methods, uh, looking at neural network analysis, time series information. So next steps we'd love to have an additional conversation with you. You can also find on some additional information around how called areas working in federal government, by going to cloudera.com solutions slash public sector. And we welcome scheduling a meeting with you again, thank you for joining us today. Uh, we greatly appreciate your time and look forward to future conversations. Thank you.

>>from around the globe. It's the Cube with digital coverage of enterprise data automation an event Siri's brought to you by Iot. Tahoe. >>Okay, we're back. Welcome back to data Automated. A J ahora is CEO of I o Ta ho, JJ. Good to see you. How have things in London? >>Big thing. Well, thinking well, where we're making progress, I could see you hope you're doing well and pleasure being back here on the Cube. >>Yeah, it's always great to talk to. You were talking enterprise data automation. As you know, with within our community, we've been pounding the whole data ops conversation. Little different, though. We're gonna We're gonna dig into that a little bit. But let's start with a J how you've seen the response to Covert and I'm especially interested in the role that data has played in this pandemic. >>Yeah, absolutely. I think everyone's adapting both essentially, um, and and in business, the customers that I speak to on day in, day out that we partner with, um they're busy adapting their businesses to serve their customers. It's very much a game of and showing the week and serve our customers to help their customers um, you know, the adaptation that's happening here is, um, trying to be more agile, kind of the most flexible. Um, a lot of pressure on data. A lot of demand on data and to deliver more value to the business, too. Serve that customer. >>Yeah. I mean, data machine intelligence and cloud, or really three huge factors that have helped organizations in this pandemic. And, you know, the machine intelligence or AI piece? That's what automation is all about. How do you see automation helping organizations evolve maybe faster than they thought they might have to >>Sure. I think the necessity of these times, um, there's there's a says a lot of demand doing something with data data. Uh huh. A lot of a lot of businesses talk about being data driven. Um, so interesting. I sort of look behind that when we work with our customers, and it's all about the customer. You know, the mic is cios invested shareholders. The common theme here is the customer. That customer experience starts and ends with data being able to move from a point that is reacting. So what the customer is expecting and taking it to that step forward where you can be proactive to serve what that customer's expectation to and that's definitely come alive now with they, um, the current time. >>Yes. So, as I said, we've been talking about data ops a lot. The idea being Dev Ops applied to the data pipeline. But talk about enterprise data automation. What is it to you and how is it different from data off? >>Yeah, Great question. Thank you. I am. I think we're all familiar with felt more more awareness around. So as it's applied, Teoh, uh, processes methodologies that have become more mature of the past five years around devil that managing change, managing an application, life cycles, managing software development data about, you know, has been great. But breaking down those silos between different roles functions and bringing people together to collaborate. Andi, you know, we definitely see that those tools, those methodologies, those processes, that kind of thinking, um, landing itself to data with data is exciting. We're excited about that, Andi shifting the focus from being I t versus business users to you know who are the data producers. And here the data consumers in a lot of cases, it concert in many different lines of business. So in data role, those methods those tools and processes well we look to do is build on top of that with data automation. It's the is the nuts and bolts of the the algorithms, the models behind machine learning that the functions. That's where we investors our R and D and bringing that in to build on top of the the methods, the ways of thinking that break down those silos on injecting that automation into the business processes that are going to drive a business to serve its customers. It's, um, a layer beyond Dev ops data ops. They can get to that point where well, I think about it is, Is the automation behind the automation we can take? I'll give you an example. Okay, a bank where we did a lot of work to do make move them into accelerating that digital transformation. And what we're finding is that as we're able to automate the jobs related to data a managing that data and serving that data that's going into them as a business automating their processes for their customer. Um, so it's it's definitely having a compound effect. >>Yeah, I mean I think that you did. Data ops for a lot of people is somewhat new to the whole Dev Ops. The data ops thing is is good and it's a nice framework. Good methodology. There is obviously a level of automation in there and collaboration across different roles. But it sounds like you're talking about so supercharging it, if you will, the automation behind the automation. You know, I think organizations talk about being data driven. You hear that? They have thrown around a lot of times. People sit back and say, We don't make decisions without data. Okay? But really, being data driven is there's a lot of aspects there. There's cultural, but it's also putting data at the core of your organization, understanding how it effects monetization. And, as you know, well, silos have been built up, whether it's through M and a, you know, data sprawl outside data sources. So I'm interested in your thoughts on what data driven means and specifically Hi, how Iot Tahoe plays >>there. Yeah, I'm sure we'll be happy. That look that three David, we've We've come a long way in the last four years. We started out with automating some of those simple, um, to codify. Um, I have a high impact on organization across the data, a data warehouse. There's data related tasks that classify data on and a lot of our original pattern. Senai people value that were built up is is very much around. They're automating, classifying data across different sources and then going out to so that for some purpose originally, you know, some of those simpler I'm challenges that we have. Ah, custom itself, um, around data privacy. You know, I've got a huge data lake here. I'm a telecoms business. I've got millions of six subscribers. Um, quite often the chief data office challenges. How do I cover the operational risk? Where, um, I got so much data I need to simplify my approach to automating, classifying that data. Recent is you can't do that manually. We can for people at it. And the the scale of that is is prohibitive, right? Often, if you had to do it manually by the time you got a good picture of it, it's already out of date. Then, starting with those those simple challenges that we've been able to address, we're then going on and build on that to say, What else do we serve? What else do we serve? The chief data officer, Chief marketing officer on the CFO. Within these times, um, where those decision makers are looking for having a lot of choices in the platform options that they say that the tooling they're very much looking for We're that Swiss army. Not being able to do one thing really well is is great, but more more. Where that cost pressure challenge is coming in is about how do we, um, offer more across the organization, bring in those business lines of business activities that depend on data to not just with a T. Okay, >>so we like the cube. Sometimes we like to talk about Okay, what is it? And then how does it work? And what's the business impact? We kind of covered what it is but love to get into the tech a little bit in terms of how it works. And I think we have a graphic here that gets into that a little bit. So, guys, if you bring that up, I wonder if you could tell us and what is the secret sauce behind Iot Tahoe? And if you could take us through this slot. >>Sure. I mean, right there in the middle that the heart of what we do It is the intellectual property. Yeah, that was built up over time. That takes from Petra genius data sources Your Oracle relational database, your your mainframe. If they lay in increasingly AP eyes and devices that produce data and that creates the ability to automatically discover that data, classify that data after it's classified them have the ability to form relationships across those different, uh, source systems, silos, different lines of business. And once we've automated that that we can start to do some cool things that just puts a contact and meaning around that data. So it's moving it now from bringing data driven on increasingly well. We have really smile, right people in our customer organizations you want do some of those advanced knowledge tasks, data scientists and, uh, quants in some of the banks that we work with. The the onus is on, then, putting everything we've done there with automation, pacifying it, relationship, understanding that equality policies that you apply to that data. I'm putting it in context once you've got the ability to power. A a professional is using data, um, to be able to put that data and contacts and search across the entire enterprise estate. Then then they can start to do some exciting things and piece together the tapestry that fabric across that different systems could be crm air P system such as s AP on some of the newer cloud databases that we work with. Snowflake is a great Well, >>yes. So this is you're describing sort of one of the one of the reasons why there's so many stove pipes and organizations because data is gonna locked in the silos of applications. I also want to point out, you know, previously to do discovery to do that classification that you talked about form those relationship to glean context from data. A lot of that, if not most of that in some cases all that would have been manual. And of course, it's out of date so quickly. Nobody wants to do it because it's so hard. So this again is where automation comes into the the the to the idea of really becoming data driven. >>Sure. I mean the the efforts. If we if I look back, maybe five years ago, we had a prevalence of daily technologies at the cutting edge. Those have said converging me to some of these cloud platforms. So we work with Google and AWS, and I think very much is, as you said it, those manual attempts to try and grasp. But it is such a complex challenge at scale. I quickly runs out of steam because once, um, once you've got your hat, once you've got your fingers on the details Oh, um, what's what's in your data estate? It's changed, you know, you've onboard a new customer. You signed up a new partner, Um, customer has no adopted a new product that you just Lawrence and there that that slew of data it's keeps coming. So it's keeping pace with that. The only answer really is is some form of automation. And what we found is if we can tie automation with what I said before the expertise the, um, the subject matter expertise that sometimes goes back many years within an organization's people that augmentation between machine learning ai on and on that knowledge that sits within inside the organization really tends to involve a lot of value in data? >>Yes, So you know Well, a J you can't be is a smaller company, all things to all people. So your ecosystem is critical. You working with AWS? You're working with Google. You got red hat. IBM is as partners. What is attracting those folks to your ecosystem and give us your thoughts on the importance of ecosystem? >>Yeah, that's that's fundamental. So I mean, when I caimans, we tell her here is the CEO of one of the, um, trends that I wanted us to to be part of was being open, having an open architecture that allowed one thing that was nice to my heart, which is as a CEO, um, a C I O where you've got a budget vision and you've already made investments into your organization, and some of those are pretty long term bets. They should be going out 5 10 years, sometimes with CRM system training up your people, getting everybody working together around a common business platform. What I wanted to ensure is that we could openly like it using ap eyes that were available, the love that some investment on the cost that has already gone into managing in organizations I t. But business users to before So part of the reason why we've been able to be successful with, um, the partners like Google AWS and increasingly, a number of technology players. That red hat mongo DB is another one where we're doing a lot of good work with, um, and snowflake here is, um it's those investments have been made by the organizations that are our customers, and we want to make sure we're adding to that, and they're leveraging the value that they've already committed to. >>Okay, so we've talked about kind of what it is and how it works, and I want to get into the business impact. I would say what I would be looking for from from this would be Can you help me lower my operational risk? I've got I've got tasks that I do many year sequential, some who are in parallel. But can you reduce my time to task? And can you help me reduce the labor intensity and ultimately, my labor costs? And I put those resources elsewhere, and ultimately, I want to reduce the end and cycle time because that is going to drive Telephone number R. A. Y So, um, I missing anything? Can you do those things? And maybe you could give us some examples of the tiara y and the business impact. >>Yeah. I mean, the r a y David is is built upon on three things that I mentioned is a combination off leveraging the existing investment with the existing state, whether that's home, Microsoft, Azure or AWS or Google IBM. And I'm putting that to work because, yeah, the customers that we work with have had made those choices. On top of that, it's, um, is ensuring that we have you got the automation that is working right down to the level off data, a column level or the file level so we don't do with meta data. It is being very specific to be at the most granular level. So as we've grown our processes and on the automation, gasification tagging, applying policies from across different compliance and regulatory needs, that an organization has to the data, everything that then happens downstream from that is ready to serve a business outcome. It could be a customer who wants that experience on a mobile device. A tablet oh, face to face within, within the store. I mean game. Would you provision the right data and enable our customers do that? But their customers, with the right data that they can trust at the right time, just in that real time moment where decision or an action is being expected? That's, um, that's driving the r a y two b in some cases, 20 x but and that's that's really satisfying to see that that kind of impact it is taking years down to months and in many cases, months of work down to days. In some cases, our is the time to value. I'm I'm impressed with how quickly out of the box with very little training a customer and think about, too. And you speak just such a search. They discovery knowledge graph on DM. I don't find duplicates. Onda Redundant data right off the bat within hours. >>Well, it's why investors are interested in this space. I mean, they're looking for a big, total available market. They're looking for a significant return. 10 X is you gotta have 10 x 20 x is better. So so that's exciting and obviously strong management and a strong team. I want to ask you about people and culture. So you got people process technology we've seen with this pandemic that processes you know are really unpredictable. And the technology has to be able to adapt to any process, not the reverse. You can't force your process into some static software, so that's very, very important. But the end of the day you got to get people on board. So I wonder if you could talk about this notion of culture and a data driven culture. >>Yeah, that's that's so important. I mean, current times is forcing the necessity of the moment to adapt. But as we start to work their way through these changes on adapt ah, what with our customers, But that is changing economic times. What? What we're saying here is the ability >>to I >>have, um, the technology Cartman, in a really smart way, what those business uses an I T knowledge workers are looking to achieve together. So I'll give you an example. We have quite often with the data operations teams in the companies that we, um, partnering with, um, I have a lot of inbound enquiries on the day to day level. I really need this set of data they think it can help my data scientists run a particular model? Or that what would happen if we combine these two different silence of data and gets the Richmond going now, those requests you can, sometimes weeks to to realize what we've been able to do with the power is to get those answers being addressed by the business users themselves. And now, without without customers, they're coming to the data. And I t folks saying, Hey, I've now built something in the development environment. Why don't we see how that can scale up with these sets of data? I don't need terabytes of it. I know exactly the columns and the feet in the data that I'm going to use on that gets seller wasted in time, um, angle to innovate. >>Well, that's huge. I mean, the whole notion of self service and the lines of business actually feeling like they have ownership of the data as opposed to, you know, I t or some technology group owning the data because then you've got data quality issues or if it doesn't line up there their agenda, you're gonna get a lot of finger pointing. So so that is a really important. You know a piece of it. I'll give you last word A J. Your final thoughts, if you would. >>Yeah, we're excited to be the only path. And I think we've built great customer examples here where we're having a real impact in in a really fast pace, whether it helping them migrate to the cloud, helping the bean up their legacy, Data lake on and write off there. Now the conversation is around data quality as more of the applications that we enable to a more efficiently could be data are be a very robotic process automation along the AP, eyes that are now available in the cloud platforms. A lot of those they're dependent on data quality on and being able to automate. So business users, um, to take accountability off being able to so look at the trend of their data quality over time and get the signals is is really driving trust. And that trust in data is helping in time. Um, the I T teams, the data operations team, with do more and more quickly that comes back to culture being out, supply this technology in such a way that it's visual insensitive. Andi. How being? Just like Dev Ops tests with with a tty Dave drops putting intelligence in at the data level to drive that collaboration. We're excited, >>you know? You remind me of something. I lied. I don't want to go yet. It's OK, so I know we're tight on time, but you mentioned migration to the cloud. And I'm thinking about conversation with Paula from Webster Webster. Bank migrations. Migrations are, you know, they're they're a nasty word for for organizations. So our and we saw this with Webster. How are you able to help minimize the migration pain and and why is that something that you guys are good at? >>Yeah. I mean, there were many large, successful companies that we've worked with. What's There's a great example where, you know, I'd like to give you the analogy where, um, you've got a lot of people in your teams if you're running a business as a CEO on this bit like a living living grade. But imagine if those different parts of your brain we're not connected, that with, um, so diminish how you're able to perform. So what we're seeing, particularly with migration, is where banks retailers. Manufacturers have grown over the last 10 years through acquisition on through different initiatives, too. Um, drive customer value that sprawl in their data estate hasn't been fully dealt with. It sometimes been a good thing, too. Leave whatever you're fired off the agent incent you a side by side with that legacy mainframe on your oracle, happy and what we're able to do very quickly with that migration challenges shine a light on all the different parts. Oh, data application at the column level or higher level if it's a day late and show an enterprise architect a CDO how everything's connected, where they may not be any documentation. The bright people that created some of those systems long since moved on or retired or been promoted into so in the rose on within days, being out to automatically generate Anke refreshed the states of that data across that man's game on and put it into context, then allows you to look at a migration from a confidence that you did it with the back rather than what we've often seen in the past is teams of consultant and business analysts. Data around this spend months getting an approximation and and a good idea of what it could be in the current state and try their very best to map that to the future Target state. Now, without all hoping out, run those processes within hours of getting started on, um well, that picture visualize that picture and bring it to life. You know, the Yarra. Why, that's off the bat with finding data that should have been deleted data that was copies off on and being able to allow the architect whether it's we're working on gcb or migration to any other clouds such as AWS or a multi cloud landscape right now with yeah, >>that visibility is key. Teoh sort of reducing operational risks, giving people confidence that they can move forward and being able to do that and update that on an ongoing basis, that means you can scale a J. Thanks so much for coming on the Cube and sharing your insights and your experience is great to have >>you. Thank you, David. Look towards smoking in. >>Alright, keep it right there, everybody. We're here with data automated on the Cube. This is Dave Volante and we'll be right back. Short break. >>Yeah, yeah, yeah, yeah

>> Announcer: Live from Orlando, Florida it's the CUBE, covering Grace Hopper's Celebration of Women in Computing, brought to you by SiliconANGLE Media. >> Welcome back to theCUBE's coverage of the Grace Hopper Conference here in Orlando, Florida. I'm your host, Rebecca Knight. We're joined by Pamela Rice. She is the Head of Technology, Strategy, and Innovation Labs Engineering at Capital One. So, thanks so much for joining us. >> Absolutely, it's good to be here. It's really good. >> So, you're only been at Capital One for nine months. >> That's right. >> You're new to the job, new to the company. Tell our viewers a little bit about why you chose Capital One and what your first experiences were. >> Yeah, absolutely. So, I'm leading technology, strategy, and innovation, and what that means is looking at how we can use emerging technologies to set our course for really creating better and different changing products for our customers. We know that customers' expectations are changing drastically and technology is really rapidly accelerating. So, using things like IOT, machine-learning, streaming data, these are all ways we can connect with our customers better. It's funny, when I first walked in the Capital One doors to interview nine months back, they had been talking to me for some time and I'm like, Okay, I'll talk to you. I didn't have high expectations-- >> Because it's a-- >> It's a bank! >> You thought it was a buttoned up kind of-- >> I had been part of Fintech and startups and I'm like, rapid, let's change the world kind of thing, and I walked into the building in Richmond, and I asked a person, 'cause I could immediately see that teams were working on things together, there was a sense of real purpose and energy. Like you could feel the energy in the walls. They were using buzzwords like machine-learning, and I was like, wait a minute. I must have gone to the wrong place. I went back to the front desk and I'm like, "Is this Capital One?" I really did that, because it felt like a tech company. The walls were bright and shiny and people were running around, and it was a little bit of a hectic pace to it. I just thought immediately, there's something to this place that looks different than my expectations of a bank were. >> What is Capital One doing right in the sense of bringing this technology buzz into a financial institution? >> I think that it's safe to say that, to Capital One, really understands that customer expectations are changing so fast, and technology is changing so fast, that they have to invest heavily in technology to really reach their customers and go where their customers are going to want to go. So, part of my job is looking at strategy. Where do we think technology is going to go and how do we make big bets now so that we can meet our customer needs in the future? I think one of the things that holds true for me is that our internal moral compass, our fiber, our DNA, comes out when we talk about how we get there. We want to get there together with diverse teams, and that's why it's so important for us to be here. Because, the way we build products is the way that we will connect with our customers, and we want our customers, all of our customers, to feel connected to our products, not just one segment of the community. >> So, talk to me a little bit about the process of designing a new product for your customers, with that diverse team behind it. >> It's funny, because in the old waterfall ways, an idea would be born in some boardroom or some shadow room, and then all a sudden, engineers would get a speck on their plate that would say, hey, go build this. Then, oftentimes, they were so disconnected from the customers. We've taken that and pivoted it totally around, so that we have whole teams that are filled with designers, with design thinking, with product, with engineering, and they can connect with the customers in a way that really optimized towards getting a product out to market faster. By the way, that tight life cycle from the time an idea is born to the time you can get something in customers' hands, that needs to shorten as well, so that you can focus on innovation. If your day-to-day activities takes you months, if not years, to get out to market, you have no time for innovation. So, part of my job also is optimizing how we get things out to market faster and in a streamlined way, and empower that team to connect with the customers better. >> And see what the customers like and what they don't like, and then iterate. >> Exactly. >> What are some of the most exciting things you're working on now? >> Well, it's hard to say that there's just one thing. We have so many really big milestones underway. It's no secret that we have a heavy migration to the cloud. We were one of the first financial institutions that really doubled down and said we're all in, we're going there. In fact, I think there's going to be more and more financial institutions that start going there, because it's such a time-saving. If I don't have to worry about all these racks and all this hardware, I spend more time connecting with the customers. Again, operational efficiency in connecting with your customers. We have a high dev-ops mindset, meaning we disconnect that bridge from engineer to throwing something over the wall, and we combine the team that has everything it takes to get something out to the customer, and then fix something if something goes bump in the night. We also have a high culture, that is we have a goal of getting into microservices more. So, instead of these big monolithic applications, really focusing on these small, microelements that have functionality that is enabled to the customer so we get things out the door faster. >> So, when you talk about diverse teams, and you're talking about a lot of cross-functional teams, so you've got teams with engineers and designers on them, working together. But, you're also talking about racial diversity and gender diversity. So, how do you make that happen? >> It's a tough problem. I've been asked a lot of questions about how do you have diversity programs that actually work, and I will say it is not just one program. When you're focusing on diversity, you can't just think about it from a program perspective. You have to think about it at your DNA level. Like every conversation, every way that you think about who should be promoted or who should get an opportunity, or economic parity, all of these things you should be questioning, am I thinking about this through a diversity lens? So, even in conversations I have with my team where somebody should be bring up something as innocuous as man-hours, I correct them. I say, it's not man-hours. These are people-hours. So, even if you can have those small hints, you need a program, absolutely. But, you need diversity included in every conversation you have, whether it is about who's going to get promoted, who's going to get a bonus, or how we talk about people and where they spend their time. >> As a woman engineer- I hate saying a woman eng- You're an engineer! You're a human being engineer. But, you also are a role model to the younger women here at Grace Hopper. What is your best advice for them, if they want to have a career as an engineer? >> It was funny, yesterday I was on a panel. Over 700 people came. I was just so honored to be part of that experience. My role in technology and being an executive now in technology really has shifted quite a bit. I feel like it's my job to give back to the community. There's nothing more empowering for me personally than to see somebody helped by words of advice or being connected to somebody else. I think that my biggest words of advice are really to know that you deserve it, know that you deserve this career, know that you deserve to dream big. You deserve a loud voice. You deserve a seat at the table. You deserve the whole table. You deserve whatever you want to dream and if you have voices in your head or external voices that are telling you you can't have it, quiet those voices and believe in yourself. Because, there's nothing more powerful than believing in yourself. We are all here believing in you, because these engineers deserve it. I would just tell them to believe that. >> That's great advice. >> Thank you. >> Well, Pamela, thanks so much for being on our show. It's been really fun talking to you. >> Thank you so much. >> We will have more from Orlando and the Grace Hopper Conference just after this. (light, electronic music)

>> Announcer: Live from Las Vegas, Nevada it's theCube. Covering, Accelerate 2017. Brought to you by Fortinet. Now, here are your hosts, Lisa Martin, and Peter Buress. (tech music) >> Lisa: Hi, welcome back to theCube. We are Silicon Angle's Flagship Program, where we go out to the events and extract the signal to the noise, bringing it directly to you. Today, we are in beautiful Las Vegas with Fortinet. It's their Accelerate 2017 Event. I'm your host, Lisa Martin, joined by my cohost, Peter Buress. And we're very excited to be joined by a Technology Alliance Partner, Nozomi Networks, Edgard Capdevielle. You are the CEO? >> Yes, that's right. >> And, welcome to theCube. >> Thank you, happy to be here. >> So, a couple of great things that Nozomi announced, just a couple of months ago, one was, they just secured fantastic $7.5 million in Series A Funding. And the second thing they announced was you, as the new CEO, so congratulations on your new post. >> Thank you very much, thank you. >> So, Nozomi is focused on the Industrial Control Systems Industry. What was it about this particular opportunity, that attracted you to want to lead Nozomi? >> Yeah, great question. Two things mainly. One, is the team. The two founders are truly rock stars, they have a great background in Cyber Security, and how do we apply Artificial Intelligence to Industrial Cyber Security. And two was, I had been working with the founders for a little bit, and I saw, with my own eyes, how the customers adopted the technology, how easy it was to deploy in an industrial setting, which tends to have a lot of friction. Not a lot of equipment gets into those networks. And the ease of proof of concepts, I saw it with my own eyes. And the frictionless interactions, made me join. >> So Nozomi was started in 2013, you're already monitoring over 50,000 industrial installations. >> That's right. >> Some of the themes that we've talked about, at the event today, so far, with Fortinet's senior leaders, is the evolution of security, where they're positioning, really at this third generation of that. As we're seeing that, and we're seeing that in order for businesses to digitalize successfully, they have to have trust in that data. What is Nozomi seeing, in terms of your industrial customers? What are some of the biggest concerns that they have, regarding security? And how are you working with Fortinet, to help mitigate or limit damage from cyber attacks? >> A lot of our customers in our space, are going through what's called IT/OT Conversions. OT networks, have traditionally been serial, point to point, run over two step para copper and they've recently adopted ethernet. When you adopt ethernet, you have a gravitational force, which is to connect. So these OT networks used to be air gaps, segregated, and now they're being converged with IT technology, under sometimes, IT operation. And therefore, they start suffering the traditional IT attacks. Those traditional IT attacks, are particularly harmful when it comes to industrial, critical infrastructure. And they require special technology that understands those protocols, to be able to detect anomalies, and white list or black list, certain activities. >> Give some example, of an IOT network. So, what is, you say critical infrastructure, gives us some examples, what are we talking about? >> IOT's a very broad term. We focus very specifically on industrial IOT. >> Or, industrial IOT. >> Industrial IOT, could be a network that controls a refining, so the refining process in a refinery. It could be electrical distribution, any form of electrical generation, oil and gas, upstream or downstream. Manufacturing, everything that moves in manufacturing, is controlled by an industrial control networks. Pharma, in the same subsegment, if you will. Some transportation, we're based in San Francisco, so our barge system is controlled with industrial control systems. >> So, we're talking about, as you say critical infrastructure, we're talking about things that, where getting control of some element of that critical infrastructure, >> Correct. >> Especially in the process manufacturing businesses, can have enormously harmful effects? >> Correct. >> On not only business, but an entire community? >> The disruption that it can cause is tremendous. From lights out in a city, to harm to people, in a transportation case, oil and gas case. Environmental damage, leakage. The damage can be tremendous. And that's basically, one of the huge differences between IT and OT. In IT, if your network blinks, your email may be two seconds late, my print job may need to be resent. In OT, you may not be able to turn off that valve, or stop this process from happening, or receive an alarm in time. >> Right, so like, I live in Palo Alto. Not too from me is, some of the big refineries up in Richmond, California. And not too long ago, they had an OT outage, and it led to nearly a billion dollars worth of damage, to that plant, and to the local environment. >> Correct. >> So this is real serious stuff. >> So with a product like Nozomi, you can detect anomalies. Anomalies come in three flavors. One could be equipment damage, malfunction. The other one could be human error, which is very very common. And the other one could be cyber. Any one of those could be an anomaly, and if it tries to throw the process into a critical state, we would detect that, and that's where ... >> Talking about cyber, from a cyber attack perspective, what is it about industrial control systems that makes them such a target? >> Yeah. It is that they had been used to be isolated networks, just like I said. When IT and OT converges, are taking networks that used to be serial security was not really a concern, in industrial control networks, you don't really have identity, you don't have authentication. You're just starting to have encryption. Basically, if you drop a command in the network, that command will get executed. So, it's about the vulnerability of those. >> Vulnerability, maybe it's an easy target? And then from a proliferation perspective, we mentioned the evolution of security. But, the evolution of cyber attacks, the threat surface is increasing. What is the potential, give us some examples, some real world examples, of the proliferation that a cyber attack, >> That is a great question. >> And an industrial control system, can have on a retailer or a bank, energy company? >> The industry was put in the map in 2010, with Stuxnet. Stuxnet was the first attack, everybody talked about Stuxnet for a while. And it was very hard to create a market out of that, because it was done really by a nation's state, and it was done like once. Since then, 2010, 2013 and now 'til today, attacks have increased in frequency dramatically, and in use cases. Not only are nation states attacking each other, like in the case now of the Ukraine, but now you have traditional security use cases, your malicious insider, you're compromised insiders, doing industrial cyber attacks. In 2015, the Department of Homeland Security reported 295, industrial cyber attacks, in our nation's critical infrastructure. And those are not mandated, they don't have a reporting mandate, so those are voluntary reports. >> Wow. >> So that number, could be two or three times as big. If you think about it, from 2010, we've gone from once a year, to 2015 once per day. So, it's happening. It's happening all the time. And it's increasing not only in frequency, but in sophistication. >> So, it's 295 reported. But there's a bunch of unreported, >> Correct. >> That we know about, and then there's a bunch that we don't know about? >> Correct. >> So, you're talking about potentially thousands of efforts? And you're trying with Fortinet and others, to bring technology, as well as, a set of best practices and thought leadership, for how to mitigate those problems? >> That's right. With Fortinet, we have a very comprehensive solution. We basically combine Fortinet's sophistication or robustness from a cyber security platform, with Nozomi's industrial knowledge. Really, we provide anomaly detection, we detect, like I said, any sort of anomaly, when it comes to error, cyber, or malfunction. And we feed it to Fortinet. Fortinet can be our enforcement arm if you will, to stop, quarantine, block, cyber attacks. >> So, Nozomi's building models, based on your expertise of how industrial IOT works, >> That's right. >> And you're deploying those models with clients, but integrating the back into the Fortinet sandbox, and other types of places. So, when problems are identified, it immediately gets published, communicated to Fortinet, and then all Fortinet customers get visibility into some of those problems? >> We connect with Fortinet in two ways. One, is we have 40 SIM, so we alert everybody. We become part of the information, security information environment. But we also used Nozomi Fortigates, to block, to become active in the network. Our product is 100% passive. We have to be passive to be friendly deployed in industrial networks. But, for the level of attack or the level or risk is very high, you can actually configure Fortinet to receive a command from Fortinet, and from Nozomi, and actually block or quarantine a particular contaminated node, or something like that. Does that make sense? >> Oh, totally. Makes 100%, because as you said, so you let Fortinet do the active work, of actually saying yes or no, something can or cannot happen, based on the output of your models? >> That's right. Yep. >> So, when you think about IOT, or industrial IOT, there's an enormous amount of investment being made of turning all these analog feeds, into digital signals, that then can be modeled. Tell us a little bit about how your customers are altering their perspective on, what analog information needs to be captured, so that your models can get smarter and smarter, and better and better at predicting and anticipating and stopping problems. >> When it comes to industrial models, you need to pretty much capture all the data. So, we size the deployment of our product based on the number of nodes or PLC's that exist in an industrial network. We have designed our product to scale, so the more information or the more number of nodes, the better our models are going to be, and our products will scale to build those models. But, capturing all the data is required. Not only capturing, but parsing all the data, and extracting the insides and the correlations between all the data, is a requirement for us to have the accuracy in anomaly detection that we have. >> What is the customer looking at in terms of going along that, that seems like an arduous task, a journey. What does, and you don't have to give us a customer name, but what does that journey look like, working together with Nozomi, and Fortinet, to facilitate that transformation, from analog to digital, if all the information is critical? >> That transformation is happening already. A lot of these industrial networks are already working on top of ethernet, a standard DCPIP. The way the journey works for us, is we provide, as soon as we show up, an immediate amount of visibility. These networks don't have the same tool sets from a visibility and asset management perspective that IT networks have. So, the first value add is visibility. We capture an incredible amount of information. And the first and best way to deploy it initially, is with, let me look at my network, understand how many PLC's do I have, how the segmentation should be properly done. And then, during all this time, our model building is happening, we're learning about the physical process and about the network. After we've done with the learning our system, determines that now it's ready to enforce, or detect anomalies, and we become at that point, active in anomaly detection. At that point, the customer may connect us with Fortinet, and we may be able to enforce quarantine activities, or blocking activities, if the problem requires it. >> Is there any one particular, use case that sticks out in your mind, as a considerable attack, that Nozomi has helped to stop? >> We obviously can't name any one in particular, but when it comes to defending yourself against cyber criminals, we have defended companies against malicious insiders. Sometimes, an employee didn't like how something may have happened, with them or with somebody else, and that person leaves the company, but nobody removed their industrial credentials. And they decide to do something harmful, and it's very hard. Industrial malicious insider activity, is extremely hard to pinpoint, extremely hard to troubleshoot. Industrial issues in general, are very hard to troubleshoot. So, one of the things that Nozomi adds a lot of value is, is allowing troubleshooting from the keyboard, without eliminating trucks and excel sheets, you quickly can pinpoint a problem, and stop the bad things before they happen. >> One more quick question for you. With the announcements that Fortinet has made today, regarding, you mentioned some of the products, what are you looking forward to most in 2017, in terms of being able to take it to the next level with your customers? To help them, help themselves? >> Listen, the solution works amazingly well. We have to tell more people about it. I think the critical infrastructure has not had the attention in prior years, and I think this year's going to be a year where, ICS security is going to be, and Fortinet of course, is very aware of this, is going to be a lot more relevant for a lot more people. The number of attacks, and the you know, the attacks surface that will never be, it's all playing so that, this year's going to be a big year. >> Yeah, I think we were talking, before we started, that the U.S. Department of Homeland Security, has just identified the U.S. Election System, as a critical infrastructure. >> That's right. >> So maybe it's going to take more visible things, that have global implications, to really help move this forward. >> I think the one point I would make when it comes to government, government has been great, if you make an analogy, this is an analogy that I have on the top of my head, if you look at cars in the automotive industry, seat belts and airbags have saved a lot of lives. We don't have that in industrial cyber security. And we need the government to tell us, what are the seat belts? And what are the minimum set of requirements that are electrical, infrastructures should be able to sustain? And that way, it makes the job easier for a lot of us, because nobody can tell you today, how much security to invest, and what's the mix of security solutions that you should have. And therefore, in the places where you don't have a lot of investment, you don't have none. And you become very vulnerable. Today, if you want to ship a car, and you want your car to be driven on the road, it has to have airbags, and it has to have seat belts, and that makes it a minimum bar for proper operation, if you will. >> But the proper, the way it typically works, is government is going to turn to folks like yourself, to help advise and deliver visibility, into what should be the appropriate statements about regulation, and what needs to be in place. So, it's going to be interesting because you and companies like you, will in fact be able to generate much of the data, that will lead to hopefully, less ambiguous types of regulations. >> Yes, that's right. That's right. I agree 100%. >> Wow, it's an exciting prospect. Edgard Capdevielle, thank you so much. CEO of Nozomi Networks, it's been a pleasure to have you on the program today. >> Thank you. >> On behalf of my cohost Peter Buress, Peter, thank you. We thank you for watching theCube, but stick around, we've got some more up, so stay tuned. (tech music)

live from Las Vegas it's the cute covering knowledge 16 brought to you by service now hear your host dave vellante and Jeff Frick we're back this is the cube silicon angles flagship production we go out to the events we extract the signal from the noise the signal here at servicenow knowledge 16 is the extension of service management across the enterprise Brian Andrews is here is a vice president of IT it's stone brewing cube alum bride great to see you again thank you it's great to be here nice to see you guys another knowledge you know I thought happened a good energy this year yeah you know I spent third knowledge how's this week been for you oh it's a blast yeah incredible energy and growth and excitement from the company the partners it's been fun so third nology that service now for two years yeah right and so what the first knowledge was sort of come and kicking yeah exactly talking all the customers is this stuff Rio exactly last year we got to speak and this year were in the customer showcase which is new one of four and telling our story about what we did and meeting other customers and partners it's fun so give us the update what's the story um you guys are growing yeah yeah so stone brewing we're the 10th largest craft beer company in the country and growing double-digit growth so yeah we're now opening a second brewery in Richmond Virginia and a third in Berlin Germany doing two at the same time which is pretty nuts for us it's a du Bois so it's a large focus for the company we're actually the first American craft brew company to open a brewery anywhere in Europe and to operate it we're on right to Berlin and in Germany hell with us I know right right into the fire I doubt well I talk about that business decision to go into Germany I mean beer central I know I know well crap beer starting to really take off in Europe and we were looking at sites all through Europe and really fell in love with this property in brillion it's a old gas works facility brick a neat place for garden inside it's just really a neat place but the crappier movements has a lot of energy there and we feel like that can be our European hub to brew and distribute throughout Europe so it's a great spot a great place to come visit and spend the day and enjoy the gardens and that's gonna be a lot of time we have a really a large bistro going in as well so it could be a place you want to stay and hang all day yeah girls that's right that's right and house I'm just curious we don't find you know kind of the German purity laws are there special you know Germany's a very special place to do business for a whole lot of reasons HR reasons and data privacy reasons and this that and the other from the brewery perspective you know we hear about their purity laws do you have to you have to follow those is your new animal as an American craft beer manufacturer how does that work well so most of our beer that we do the core beers they do fit right into that our stone IPA arrogant bastards they fit in but we do a lot that do not fit in because we add in espresso or tangerine or good stuff like that so we're purposely going to be knocking down that bureaucracy and being rebellious we had a event last week where we served only beers that did not comply with the law true to our culture were rebels and it's exciting for us so i have to say i mean german beer is special I'd consume a lot of German of year in my day and somehow the next day you just feel great yeah absolutely is that the experience with stone yeah yeah it's gonna be you know I think a new to get the strong you know bitter forward hot forward IPAs that we serve will be different that's awesome now you guys you find you saying brought in service now from the business side yeah first we did an NIT but you but you led that acquisition so two years ago we were looking at putting in a set of systems for the business each group had their own needs and they had selected systems they want to bring in the brew ops maintenance was number one that we needed to serve as a use case so the demand was really growing for our beer as it's been we need to keep up with the demand and so we can't have the brewing and down we were turning to a 24-7 operation and the brewery anytime a piece of equipment was down we're not getting beer to our fans I'm not serving our customers so we we needed something for planned maintenance to keep that equipment rolling facilities wanted something as well for maintaining the facilities and HVAC units and all that safety I wanted something for reporting incidents they were all all those groups were outlook and excel and so they needed a system they didn't have one we had some project management needs in our marketing group and of course I T wanted a great system too so we looked at those and said we can collapse all these down into one system with service now because in the end they had a common set of requirements they wanted workflow and reporting and visibility work order management so we did some proof of concepts and they bought in and we deployed service now to the business first because they had nothing at least I t had something it was antiquated we had something so we serve we serve them first and so you're one was all about putting those platforms in place to crawl walk and then you're too we're optimizing and now we report some with some terrific results that have come out now by using it I get a triple it and take it overseas no you go right straight to the run that's right that's right and fly haha so as you grow what role do you see service now playing I mean have you been able to sort of sense or measure the productivity impacts and we've had some great results that come out of this so our brewing department as I said they need to keep that equipment rolling any downtime was hurting us we cut the downtime in half by using planned maintenance and so we use not only the corrective work orders but planning's we have 2200 items from the brewery and packaging in our cmdb we're unplanned maintenance against those now half of the work orders that were completing our plan preventative in nature those were a very small percentage earlier it was more reactive and corrective moving to planned we're more on top of things more proactive and the equipment's up and running longer so she's meant to the CMDB so you if you've got a single cmdb or you there we do yeah single cmdb for all the brewing and packaging equipment and it's all as a nice data hierarchy so we can know that it's the escondido brewery it's brewhouse one and it's the latter ton and that has valves and pumps and sensors now those items might be used at other pieces of equipment too so we can put those assign them to different items in the CMDB but it's all in there and organized and we can you know see how we're doing on cost control and when we need to replace equipment or maintain it and on the preventative is it implementing you know suggested best practices by the manufacturer of those components or did you guys come up with your own kind of maintenance schedule based on operating experience etc yeah primarily from the maintenance from the manufacturer so we have those in is knowledge articles as well and then week but we have around procedures that we also would put in there and those those are put through in the work order so the technicians can see those and then one thing that's really nice is when we have down time in the brewery for maybe the brewing team is doing training we can see all the planned maintenance coming up and accelerate some so we may have something for next week we can move it up by a few days or something we may want to delay so we can have less downtime and group it together and do that maintenance all at once what kind of modifications have you did you have to make or did you have to make bringing in service now well we were a little on the bleeding edge in some cases a couple years ago as we were putting in the facilities maintenance and the planned maintenance so that was just starting to come out with service now so we had to build some custom tables and are we want to make sure it made sense for the the context so we had crafting assets and crafting systems those kinds of things so the business contact makes sense but those are now coming in out of the box so we're starting to pull back on the customization so it was not too bad a few things now as we excited the facilities and safety we want to make sure we could tag items if there's a leaking valve or exposed why are those kinds of things we can tag it as a facilities issue a brewing ops issue but also note it as a safety issue Safety's big it's down we're going to make sure it's a safety safe environment for our team we've cut injuries in half by having a focus on people and training the processes but also having this too well now to make all the issues visible and real-time so we're having a hundred percent increase in safety issues reported to us so we can see more they were out there before and weren't being reported or lost an email in Excel so we're seeing those now more proactive fixing and cut injuries in half we're really proud of that talk about the process behind that because we always talk about the you know the people process technology technologies one piece a fool with a tool you have blob of ulva all the little idioms but you're using service now as a platform to enter those incidents those safety incidents but somebody's got to actually do that right so then is it the person who got injured and what's the incentive for them doing that or explain the process behind that while safety is woven into our culture so we want to make sure day one everyone knows that's critical for us we want to leave as safe as healthy as you were when you started your day so what we have is that that form is available through our Service Catalog along with IT requests facilities request burry there's a safety incident you can report those come through from the team member that saw it so it could be the person that experienced it or someone who saw something and maybe they're working at the packaging line and they see something that could be an issue so those those could be sent through easily on a tablet or from their workstations and then the safety manager gets alerted to that works the q's runs the reports passes it to who's in charge it may be fixed by the facility's team or an engineer so they pass those tickets along that's a real plus for us having that on one system because originally the brewing folks wanted their system they were used to and that was different than what facilities had used before or safety in their previous companies but bringing it all together in one they can pass those tickets long and tracking a lot easier in one system then you're able to identify commonalities and an attack like they showed this morning's keynote the big red box you know that's right and so you were able to drill down into those and then try to put in new processes yeah remedi eight and of course all the categories what types of injuries are happening you can focus on the top ones you know is it slip and falls or lifting or forklift and those tie into then training and certification and getting people recertified so it starts the tide of learning management program as well but the other thing we hear over and over is that in the the implementation and execution with service and out and in department a and then it integrates over to b c and d and they start to say hey we want to do this two of every are you seeing a proliferation beyond kind of what your core initial delivery was oh absolutely yeah people are there's a conga line we like to say people waiting to get on our next is going to be project management for getting a new beer released so that's a seven month process or so to get from concept to actually getting the beer out the door and so we're going to be putting that all in service now for project management and having those tasks visible for everyone involved it's a really cross-functional effort to get a beer released and many different groups have to collaborate and making that visible in a single place single plan having dependencies in there and what we love about the project management suite is that the work being done is in the project plan but it's also the tasks that could assign the people to do the work and if they're getting production support or incidents that come through they can see that and there my work use along with their project work all in one place so we're really excited about putting in project management ago what are you using today for project management well for that new beer process that's a lot of Excel spreadsheets and email some document word docs those kinds of things but we have MS project and project server that we're using for construction projects but there's a lot of manual work that goes with that yet will you and have you when we will start with have you when you brought in service now were you able to retire some systems did you get rid of stuff well for IT we had a system it was the tracking system from BMC so that's one we wanted to replace so we're rolling it out for IT was a big win and that's now gotten pretty far wording incidents and change we'd like to get into problem and really start to mature that but we put the business first so I t's taking the backseat on resources but it's definitely we're well past where we were before so we'll be putting the assets in the database for IT as we've done for the brewing equipment and the facilities equipment and really build out IT ahead but ms project will definitely be retired as you move and most of the other ones the media department has a system they used called a sauna and they use that for project management that will also go when we have the new beer system getting launched with project management do you how do you deal with the organization or is their organizational friction people say want to hang on to the last user ah the other stretcher right right how do you deal with that ah well so most of the folks were using outlook and excel so those are pretty easy they really they needed something and didn't have it so those were easier wins but you know there's some the change management interesting because when you look in the magic quadrant you know what's the best maintenance management systems or project management systems and service now yet isn't out there right because it's the best in service management but getting people to see that it can also be a terrific system for project management or maintenance is a bit of a stretch right so you have to show them really well what is it you really need what are those requirements so let me show you so we've done some proof of concepts and that's been helpful to get people to see as well and believe because they see it as an IT system mostly when they go look it up but we've shown what we're doing and they get it it's exciting so we started last year we talked about time to value when we sort of joked time to beer right have you been able to actually quantify that do you see faster time to beer well it's like having that brewery equipment up and running has been big for us and cutting that in half of the down time we're getting the beer out the door so that has been the biggest win for us really I think with the seven-month new beer release process although cutting that time down isn't the number one driver of that it's more about getting it visible and collaboration and people working Heather I think that that's will be pleasantly surprised with how that's going to decrease so give us the road map over the next 12 12 months what do you be working on what's what's exciting you yeah so a couple of big things so we'll be doing that new beer project management we're also gonna be integrating with our ERP system so for the team that's getting those that maintenance requests in for the brewery they want to get those parts consumed from our European get the parts in we can track the total cost of the maintenance that's going on but also trigger reorders for the parts based on min values in our ERP so that'll be a nice integration will do the new beer and then we want to get IT mature through the IT Service Management and we're seeing so many great things with that performance analytics that's exciting to us because we're getting a lot of data good operational reports but we'd love to get some of that predictive business intelligence coming so those are a couple areas we're really looking at this year and I think also making take advantage of that the tools to make the user interface really nice-looking will be great so our service portal Service Catalog has a lot of great items on there but it doesn't look that great yes we're gonna make it look slick with some of the new tools and I guess helsinki's got some really good so you service now for that ui/ux yeah and yep NP you say bringing forth part parts of Helsinki yeah yeah so we're upgrading later this summer we're moving to Geneva in a couple of weeks and then we'll be really focusing later in the summer and making that service catalogue look good now stones got some beautiful imagery we have great shots of the beer and our facilities really great external when people see stone has really just terrific images and videos we want to make that look as good on the inside as it does on the outside for a fan so people come in and join the company and see how good we are on the inside too that's important to us so who does that beautification do you have a UI UX team that does that or is it just what you guys are pretty small team only 17 and IT that take care of a thousand team members so we have we're stretched pretty thin we have a terrific system administrator who also does development you and another gentleman that works on our websites so I think collaborating together and the tools that are available I think we'll be able to make it look good internally and we feel you have some great partners as well awesome yeah all right Brian listen thanks for coming back to the the cube and sharing your stories it's we love having stone brewing on any time so you'll appreciate it thank you very much guys appreciate being here logo all right I keep right there buddy right the cube would be back right after this at knowledge 16 Vegas right back every once in a while