>>Hello, everyone. This is not appropriate to lapse of America. I'm going to talk about the motivation. For zero knowledge goes back to the heart off, winding down identity, ownership, community and control. Much of photography exists today to support control communications among individuals in the one world. We also consider devices as extensions of individuals and corporations as communities. Here's hoping you're not fit in this picture. What defines the boundary off an individual is the ability to hold a secret with maybe, it says, attached to the ownership. Off some ethic, we want the ability to use the secret to prove ownership of this asset. However, giving up the secret itself essentially announced ownership since then, anybody else can do the same. Dear Knowledge gives us tools to prove ownership without revealing the secret. The notion of proving ownership off a digital object without revealing it sounds very paradoxical outside the model off. So it gives us a surprise when this motion was formalized and constructed by Goldwasser Miccoli and back off in the late eighties, we'll focus on the non interactive >>version of Siri, a knowledge our music in the >>stock, which was first developed by blow Tillman and Peggy, where the general it can span multiple rounds of communications music only allows a single message to be trusted. No, let's get into some technical details for musics. The objective of for music is to show that an object X, which you can think off as the public footprint, often asset, belonging clan and the language without revealing its witness. W, which you can think off as the Future Analytics team consists off three algorithms, video proof and very. The key generation process is executed by a trusted third party and the very opposite, resulting in a common >>random string, or steers, which is made public. The >>true vendor produces a proof by based on the CIA's X and the very fine with the checks. The proof against X and accepts or rejects music off course has to satisfy some properties. We needed to be correct, which basically says that when everyone follows the protocol correctly on, so we can expect, we need to be thought, which says that a false statement cannot be proven. The channel is a trickier properly to form this. How do we capture the intuition behind saying that the proof there is no knowledge of the witness. One way to capture that is to imagine their tools is the real world where the proof is calculated. Using the witness on there's a simulation worth where the proof is calculated without a witness. To make this possible, the simulator may have some extra information about the CIA's, which is independent off the objectives. The property then requires that it is not possible to effectively distinguish these words Now. It is especially challenging to construct music's compared to encryption signature schemes, in particular in signature schemes. The analog off the Hoover can use a secret, and in any case, the analog off the very fire can use a secret. But in is it's none of the crew layer and the verifier can hold a secret. Yeah, in this talk, I'm going to focus on linear subspace languages. This class is the basis of hardness. >>Assumptions like GH and deliver >>on has proved extremely useful in crypto constructions. This is how we express DD it and dealing as linear software. We will use additive notation on express the spirit logs as the near group actions on coop elements. You think the syntax we can write down Deitch on dealing Jupiter's very naturally a zoo witness sector times a constant electric so we can view the language as being penetrated by a constant language. Metrics really was hard by many groups in our instructions. What does it mean? S while uh, Standard group allows traditions and explain it off by in your group also allows one modification In such groups, we can state various in yourself facing elections. The DDN is the simplest one. It assumes that sampling a one dimensional space is indistinguishable from something full professional. The decisional linear assumption assumes the theme from tours is three dimensional spaces generalizing the sequence of Presumptions. The scaling the resumption asks to distinguish between gay damaged examples and full it and >>examples from a K plus one national space. >>Right, So I came up with a breakthrough. Is the construction in Europe 2008 in particular? There? Music for many years Off Spaces was the first efficient >>construction based on idiots and gear. Structurally, >>it consisted of two parts Our commitment to the witness Andre question proof part and going how the witness actually corresponds to the object. The number of elements in the proof is linear in the number >>of witnesses on the number of elements in the object. >>The question remains to build even shorter visits. The Sierras itself seemed to provide some scoop Rosa Russo fix. See how that works for an entire class of languages? Maybe there's a way to increase proof efficiency on the cost of having had Taylor Sierra's for each year. This is what motivates quality and after six, where we let the solace depend on the language itself. In particular, we didn't require the discrete logs of the language constants to generate this, Yes, but we did require this constant student generated from witness sample distributions. This still turns out to be sufficient for many applications. The construction achieved a perfect knowledge, which was universally in the sense that the simulator was independent. However, soundness is competition. So here's how the construction differed from roots high at a very high level, the language constants are embedded into the CIA s in such a way that the object functions as it's only so we end up not needing any separate commitment in the perfect sense. Our particular construction also needed fewer elements in the question proof, as there On the flip side, the CIA's blows up quadratic instead of constant. Let's get into the detail construction, which is actually present with this script. Let the language apparently trace by Giovanni tricks with the witness changing over time, we sat down and matrices >>D and B with appropriate damages. >>Then we construct the public series into what C. S. D is meant to be used. By the way. On it is constructed by >>multiplying the language matrix with D and being worse, Sierra's V is the part that is meant to be used by the very fair, and it is constructed using details be on be embedded in teaching. >>Now let's say you're asked to computer proof for a candidate X with fitness number we computed simply as a product of the witness with CSP. The verification of the truth is simply taking with the pairing off the candidate and the proof with the Sierras. Seeming threats is equal to zero. If you look carefully. Sierra's V essentially embedded in G to the kernel of the Matrix, owned by the language metrics here and so to speak. This is what is responsible for the correctness. The zero knowledge property is also straightforward, >>given the trapdoor matrices, D and B. Now, >>when corrected journalism relatively simple to prove proving illnesses strictly The central observation is that, given CSP, there is still enough entropy. >>India and me to >>random I seriously in particular Sierra's we Can we expand it to have an additional component with a random sample from the kernel allows it. This transformation is purely statistical. No, we essentially invented idiots are killing their talent in the era of kernel part in this transform sitting within show that an alleged proof on a bad candidate and we used to distinguish whether a subspace sample was used for a full space >>sample was used at the challenge. The need >>to have the kernel of the language in this city. That's the technical >>reason why we need the language to come from a witness. Sample. >>Uh, let's give a simple illustration >>of the system on a standard Diffie Hellman, which g one with the hardness assumption being idiot. >>So the language is defined by G one elements small D, E and F, with pupils off the phone due to the W. After that ugly, the CIA is is generated as follows example D and >>B from random on Compute Sierra speak as due to the day after the being verse and Sierra's V as G to do to do the big on day two of the video. The >>proof of the pupil >>detail that I do after the bill is computed using W. As Sierra Speed race to the party. I know that this is just a single element in the group. The verification is done by bearing the Cooper and the proof with the Sierras VMS and then checking in quality. The >>similar can easily compute the proof using trapdoors demand without knowing that what we are expecting. People leave a Peter's die and reduce the roof size, the constant under a given independent of the number of witnesses and object dimensions. Finally, at Cryptocurrency 14 we optimize the proof toe, one group >>element under the idiots. In both the works, the theorists was reduced to linear sites. The >>number of bearings needed for ratification was also industry in years. This is the crypto Ford in construction in action, the construction skeleton remains more or less the famous VR turkey. But the core observation was that many of the Sierras elements could were anomaly. Comite. While still >>maintaining some of this, these extra random items are depicted in red in this side. >>This round of combination of the Sierras elements resulted in a reduction of boat, Bruce says, as also the number of clearings required for education in Europe in 2015 kills, and we came up with a beautiful >>interpretation of skill sets based on the concept of small predictive hash functions. >>This slide is oversimplified but illustrated, wanting, uh, this system has four collecting >>puzzle pieces. The goodness of the language metrics okay again and a key Haider when >>the hidden version of the key is given publicly in the Sears. Now, when we have a good object, the pieces fit together nicely into detectable. However, when we have a bad object, the pieces no longer fit and it becomes >>infeasible to come up with convincing. Zero knowledge is demonstrable by giving the key to the simulator on observing that the key is independent of the language metrics. >>Through the years, we have extended >>enhanced not mind to be six system, especially with our collaborators, Masayuki Abby Koko Jr. Born on U. >>N. Based on your visits, we were able to construct very efficient, identity based encryption structure, resulting signatures >>public verifiable CCS, secure encryption, nine signatures, group signatures, authorities, key extremes and so on. >>It has also been gratifying to see the community make leaps and bounces ideas and also use queuing visits in practical limits. Before finishing off, I wanted to talk to you a little bit about >>some exciting activities going on Hyper ledger, which is relevant for photographers. Hyper >>Leisure is an open source community for enterprise. Great. It's hosted by the minute formation on enjoys participation from numerous industry groups. Uh, so difficult funded to efforts in Africa, we have versa, which is poised to be the crypto home for all. Blocking it and practice a platform for prospecting transactions are part of the legs on the slide here, >>we would love participation from entity inference. So >>that was a brief history of your analytics. Thanks for giving me the opportunity. And thanks for listening

>> Announcer: Live from Las Vegas, it's the Cube, covering HPE Discover 2017. Brought to you by Hewlett Packard Enterprise. >> Okay, welcome back, everyone. We are here live in Las Vegas for the Cube's exclusive coverage, three days of HPE, Hewlett Packard Enterprises Discover 2017. I'm John Furrier. My co-host Dave Vallente. Seven years of coverage, in our seventh year, and of course, we've had many guests on over those years. And our next guest has been on every year. Steven Deitch, Vice-President Worldwide Service Provider Business. Great to see you. >> Good to see you. >> Seven year Cube alumni. You've been on every year. >> That's right. >> Great to see you. >> Good, just getting older. (laughing) >> And smarter. >> Co-Host: I think we started at VM World. >> We did, way back. >> Yeah, yeah, at Barcelona, I think you were on at Barcelona. We had no live. Alot's changed. I mean, what's up with you right now? Before we get into some of the historical on where we've been and where we're going, what's happening for you in the news here at HPE Discover? What's the big story? >> Well, you know, the headline, and what Meg and Antonio and everybody else have been talking about, is HPE strategy core belief's vision, which revolves around three elements. Making hybrid IT simple. Powering the Edge and then the expertise that goes along to bring that all together. My focus is really around that hybrid IT portion. Hybrid IT is pervasive, on prem, off prem, traditional IT, private cloud, sorry, public cloud. And customers are increasingly moving to that model given the value that they see, of optimizing their IT environment and sticking workloads or sourcing applications from the best execution venue. My personal focus right now is around the service providers that will deliver the off-premise element of HPE's hybrid strategy going forward. Because we made some very clear decisions that we weren't going to do that anymore. We had a public cloud before, that we decided to shut down. With the spinoff of enterprise services, that leaves us dependent, or actually embracing partners to deliver all of that consumption-based off-premise service element. >> I mean, a lot's changed, I mean, the elephant in the room is, obviously, the decline in people buying boxes and, or hardware, peddling hardware, but IT's not declining. IT's shifting. The services model is interesting. Service provider roles are changing. You know, anyone who's in the SAAS businesses, enterprises having SAAS products that they offer their customers. In essence, a traditional enterprise buying data center hardware and software from HPE is now providing a service to their customers. >> Steve: That's right. >> With digital. >> Steve: That's right. >> This is the digital transformation. How does that shift? How do you guys talk to customers now? Because now, the service provider definition has increased. Enterprises have, maybe a portion of traditional enterprise, but also now service provider component. How do you guys talk to customers? 'Cause this is truly where the business transformation is hitting the road. How do you guys talk to customers about this trend? >> Well, let's start at, you mentioned a digital transformation. At the end of the day and in simple terms, it's entities utilizing digital technologies to improve the experience of their constituents, partners, customers, employees, processes, systems, and so forth. Hybrid IT ultimately is one of the enablers behind the digital transformation. We're extremely passionate about that because you're right, that's where everybody's going whether you're small in the market, you're mid, you're a large enterprise, or you're a service provider. You're going through your own transformation as you go forward to be able to deliver against that digital transformation process. >> You said before we're kind of reliant on, then you sort of amended that and said embracing the cloud. In fact, if you don't have a cloud strategy today, you're toast. You are relying on your partners for a big part of that strategy. It's not just Azure. It's not a one trick pony. Can you talk about sort of beyond the big partner, what you're doing to differentiate within that next tier and how they're differentiating from the big guys like AWS. >> Right, and you're absolutely right. We firmly believe the world's going to be multi-cloud so certain work loads will stay in the data center, certain will be private cloud in the data center, others will move to managed private cloud, off premise. Then others will make a lot of sense to go to a hyper-scale provider like Amazon, Azure or Google. You want the best execution venue for that application or workload. It makes all the sense in the world. That's what we call, and you've heard us talk about this before, the right mix. As customers look to where they're going to put those workloads we're working with service providers below those big hyper-scale, big gorillas, to project or deliver value that the hyper-scale providers can not. Everything is not going to go to Amazon. It's a fact. >> It's not a winner take all game. >> It's not a winner take all. The world is way too diverse. Diverse workloads, diverse geographies, diverse business requirements. The way we look at it and we embrace the service provider's below the gorillas, we want to collectively go after opportunities that the hyper-scale providers can not deliver on. It really revolves around three things that we believe, we collectively, but more importantly those service providers should be able to do. One is embrace customer complexity. Go beyond simple services. Full stack SOAs. Drive digital transformations. Embrace customer intimacy. The big gorillas, they have a very broad set of services, very rich set of services, but when it comes time to intimacy and customization, you're not going to go there. A lot of customers remember 98% of the value in the market today is still traditional apps. Number two is geography. We all know that the big boys are in about 15 or 16% physical countries today. There's still 200 countries that don't have a physical presence and when you look at data resonancy, data privacy and so forth, or even performance in latency, you still need that physical presence. Even in the countries where the big hyper-scale providers are, you still need the girth of resources. Technical, sales and so fort and sometimes that's missing. Enterprise customers and mid-market customers want to embrace that. Finally, you know this as well as everybody else, and you made that point before, as customers evolve to hybrid, they have to manage that environment. The combination of on-prem, let's call it Tier 2/Tier 3 service providers, and then the AWSs and the Azures and the Googles of the world. That's a challenge. That's a big challenge to be able to manage that hybrid environment. The service providers that we're working with, we want them to be that hybrid manager, we want them to be that broker in order to mitigate the risk, determine the best execution venue and really deal with the challenges that these guys are going, including cost, time, and skill sets. >> If I could follow up on that in terms of the sustainability of those three differentiators. Complexity, I think you're okay. I think IT just keeps getting more and more complex. The GEOs, maybe slowly over time that changes, but your point is the local resources is probably not something that the big guys are going to put in place any time soon. Belly to belly. It was interesting to hear the CEO of Wipro talk about hyper automating, but we're still decades away from eliminating all the people required. Managing multi-cloud. That seems to be a big one that is a white space right now that nobody has really cornered. >> John: Huge. >> It's not likely that any one of the businesses, Amazon's not going to own multi-cloud management. That's really not even their interest. >> John: That's single cloud. >> To me, number three is a multi-hundred billion dollar opportunity for the market and HP specifically. >> Absolutely. We go hard at all three of those and some are more defendable than others, the geography, you're absolutely right, but the resources will continue to be a challenge for folks. Number one and number three are clearly ways that our service provider partners can take advantage of opportunities that the hyper scale providers will not be able to. >> John: Why HP? >> Why HP? At the end of the day, we bring best of class technology, we bring best in class commercial models, we bring collaborative go to markets. By the way, we don't compete with our partners. I challenge folks to look at their existing vendors and ask those questions. Particularly if you're a service provider partner. Ask those questions to your existing vendors and ask them, why are you competing against me. We are very, I'll use the word clean. Strategy is very simple, very clean, we're not competing. >> John: No hair on those partnership deals. >> No hair. >> If you take out the big hyper-scalers, AWS, Google, and FaceBooks of the world, there's a big torso mid range market that you guys are going after. You didn't have competition. You're going to have all your normal competitors that we all know and talk about going after that same space. Differentiations are what you said. How are they approaching it? They're going to try and create fud around what you guys are doing and certainly this transformation market that we're in is kind of confusing. People now being more educated on the cloud which is a good thing. There's still no real definition of what multi-cloud is. Multi-cloud is happening. how are you guys competing directly with the competitors and how are you guys going to go in to win? >> I think, the fact that we're partner first and we already understand how partners function and what they need in the requirements, it sounds a little simplistic, but at the end of the day, you have a whole lot of service provider partners out there that are pure, but you have thousands of service providers and what they've done? They've evolved from being a traditional reseller or a solution provider, to adding a third business model of being a consume oriented service provider and the fact that we understand the journey that they've been on, the challenges that they go through, I will challenge our competitors to have that deep of insight that have not been channel friendly at all. >> Is that the big transformation? That third point you mentioned, is that the big change in the service provider transformation, is that that consumer focus?L >> It is because we all recognize the big service providers whether you're a big cloud service provider or a consumer service provider like an Uber or Spotify, or a Telco. Think about all of the service providers out there, let's call them, for the lack of a better word a hybrid partner. They have a resale business where they do transactions, they have a solutions business and then they have a consume business. Those are the ones that are actually capable of pulling off the differentiation. They can get intimate with the customer - >> John: They have specialism. >> They have specialism, they have professional services, they have industry insight and they understand their customers much better. >> The channel's turning into the customer for you guys in the way the partner first message - >> It's a different type of partner. Different type of partner. Absolutely. Those three swim lanes. We look at partners will either be in one, two, or all three of them. >> Steve, thanks for coming on the Cube again. Appreciate seeing you. Big takeaway from the show here, the transformations in full swing, the market's kind of going crazy with cloud and IOT. What's your big takeaway from this show this year? >> The clarity. The clarity and the focus that Hewlett Packard has and the fact that our partners and customers are really embracing it. That's the key message that I've heard from everybody. Everybody's super excited and there's a focus. I think maybe in the past, because we've been so big and so complex, but the fact of our skinning down, going in opposite directions as some of our competitors, that clarity will lead to execution excellence, I believe. >> Awesome. Stephen, thanks for taking the time. This is the Cube live coverage from HPE Discover 2017, our 7th year we're covering the transformation. More live coverage after this short break. Stay with us. We'll be right back.