Enterprise Blockchain Path to Production

[Music] are we great so we are live tania welcome to everyone we have now another exciting keynote with tanya from vmware welcome to the stage tanya we are three minutes late so don't you worry we'll have another three minutes more with tanya so tiny in the states is yours thanks so much victoria it's wonderful to be here and hello everyone good evening i'm tanya shastri i'm vp of product for blockchain at vmware and i'm very happy to be here today to talk about blockchain and the enterprise and to share our thoughts on bringing you know enterprise grade applications into production um to get started before before we start especially for those who aren't familiar with vmware i have a very brief introduction um vmware has a long heritage over 20 years of delivering innovative technology software and software as a service to our customers so that our customers in turn can innovate in their businesses and it is through this experience that we have a deep appreciation for what it takes to bring business critical applications to production and then to continue to manage them operate them and so on in production and then also here i'm just sharing a few examples of the experience that i'm drawing on to develop our viewpoint on bringing applications to production we have been partnering deeply with industry leaders in the dlt space with our product product vmware blockchain and it is through this experience of these engagements that i intend to share how leading enterprises are leveraging dlt-based solutions bringing them to production and transforming their industries so for example we have collaborated deeply with broadridge financials one of the leading financial services financial market infrastructure providers on developing the distributed ledger repo platform this is underpinned by vmware blockchain it is live it is in production um delivering over 35 billion in uh daily average volume and it continues to grow and you know they're really transforming a critical business function a critical financial market function of repo transactions and repo agreements um using this distributed ledger repo platform similarly and then we i'll just also mention um australian securities exchange their application called chess plus um is leveraging vmware blockchain and that chess plus application is is modernizing their entire post trade clearing and settlement functions and this underpins the entire australian equities market and this has gone into industry-wide testing um in november this year um and is starting to open up to the industry so that they can now start using the platform and trying out and putting their applications onto the platform another example is the symphony is their synthetic product which is a dlt as a service platform so what they're doing is they're offering blt as a service to their customers and the entire australian and new zealand markets and this is powered by vmware blockchain it's driving a completely new business revenue stream for them in addition we're also collaborating with customers like deutsche bourza bursa malaysia and others in their journeys to dematerialize um fully demoted utilize and digitize assets so these are some examples on which we're drawing um our experience and and sharing with you um what we are seeing in the industry so first off you know when we when we are talking to customers in the space it's becoming clear that there's a much better understanding that dlt provides value in a range of use cases um and all of them have somewhat of a similar problem statement right um if you look at all these um examples on the left hand side here and this is not complete it's just a an example of them whether it's bond insurance clearing settlement supply chain use cases carbon credits digital identity so on and so forth several of them either they exist today or are done in ways that are pretty outdated these multi-party systems use um architectures that have silos between the parties each party has their disparate data and processes and so on um messages need to be sent between the parties and you know when they interact and data reconciliation is required across all these parties so it's it's pretty inefficient leads to errors pretty slow multiple manual steps to recover from errors um you know increased cost and business risk right and the situation is only getting compounded because of things that you know there's increased growth demands costs are increasing competitive pressures and so on and it's quite clear that blockchain can really transform these multi-party applications with these core aspects right the shared and trusted source of truths that all parties can um can trust neutralized workflows so that processes can be automated across these multiple parties and the programmable asset aspects that allow for the assets to be more intelligent and um to actually trigger um the processes across um all parties and so that it's done in a way that that process can also be trusted at the end of the day reducing the operational cost the business risk and generating new business revenue streams so this is all great right there is a clear understanding of the value and then as as customers are looking to now bring some of these use cases into production you know this isn't enough right there's more that needs to be taken into consideration so what are some of these things that enterprises need to take into consideration as they think about taking these core primitives that are that are being delivered by blockchain and blt and then bring them into enterprise-grade solutions so here you see privacy of course is is very important for business reasons for regulatory compliance reasons privacy is always an important functionality to have performance these you know we're talking about very high volume use cases high throughput low latency requirements with you know many millions of transactions per day um efficiency is very important many of you are familiar with the inefficiency of proof-of-work mechanisms for consensus and the industry moving to proof-of-stake and so on um you'll see my background here with the with with the tree vmware or my company is very environmentally conscious and definitely you know our consensus mechanisms our efficiency is uh to move further beyond even the efficiency of proof-of-stake um capacity management um is i'd say the next one um i'll go into a little bit um as we discuss it further as well uh the predictability of costs um with gas price fluctuation and so on these things make it very difficult for enterprises to run their business so these are all considerations that need to be taken um while uh going into production right and these are the kinds of things that have to be addressed prior to going into production and above that i'll just add you know this the space is rapidly evolving there are many new things happening so thinking of the flexibility of the platform the extensibility of the platform on which these smart contracts and applications can be run um is also an important consideration and how simple and easy is it to manage and maintain because already there can be complexities when you're moving into a new space and offering new services and changing the paradigm and transforming things um how do you kind of reduce the the areas where there is complexity and then of course you know the requirement to have very extreme robustness right these are things that we don't want going down they can have impact to businesses impact to the entire ecosystem and you know having the right robustness and the right support behind these solutions um is also pretty important so well you know while i've given a quick overview of some of the things that enterprises need to consider in addition to the core blockchain capabilities i want to highlight a few here um that we come across more often and that i'd like to just bring to everyone's notice right um as important considerations for production um one is capacity management um there's there's this expectation in some ways that that blockchains and the distributed ledger technology the data that's put in is immutable and you can you know you don't take data off the blockchain but you know when we're talking about use cases where multiple terabytes of data are being put into the blockchain and where a lot of the data has to be maintained on chain um to get the the the data integrity and the verifiability and so on um we really need to have a way to prune data off the primary blockchain um but but then do it in a way that is still consistent with the core principles of blockchain so that you can maintain the full history of the blockchain in in what we call kind of a full copy client node right um so doing that well and then also at the same time ensuring that capacity management and pruning is done with consensus so that all nodes in the network have participated and it's done through the right types of governance so um capacity management is is something that needs to be thought of early and needs to be planned and designed into the system and and the infrastructure needs to be capable of doing it so i say capacity management is is an interesting one especially for high volume use cases and we are seeing several of those um already the ones i described that we're drawing our experience on are certainly in that in that category um i'll also mention upgrades so upgrades as well for all of us familiar in the in in the blockchain and dlt space upgrades are quite unique you know um the entire system has to be upgraded again with consensus and meeting the right uh primitives of the blockchain space and it has to be done with data continuity across you know there's aspects of the smart contract that may require upgrades the blockchain infrastructure that might require upgrades and how is it done in a trusted and authenticated way as well as with data continuity and then you know without having the complexity in this and making it unwieldy right because as we as we expect in the early days um given there are so there may be more features and functionality that need to be added to the smart contract or what have you upgradeability is an important consideration and again considering it early and planning and designing for it early is is something we highly recommend and then the third one i picked was reconfiguration um when we when we when we are seeing these blockchains uh networks evolving and the the market participants um participating these are growing ecosystems and market participants have to be added um scalability considerations have to be taken into into consideration you know often times to meet business goals you may want to start smaller and then grow as the market participants and the adoption increases so being able to reconfigure and scale up potentially scale down in some cases depending on the business requirements um these are the things that also have to be considered early in as a part in in the path to production and how is reconfiguration also done in a way that stays true to the blockchain primitives so that there it is done with consensus done in a trusted and authenticated fashion um and done in a way um that um that meets the blockchain primitives i think these are things that are um are important considerations to keep in mind or at least certainly some of them um and i don't want to miss um the point about having robust end-to-end support because you know in some cases there are there are several pieces coming together um there is the infrastructure side and the actual blockchain itself there is the smart contract there are aspects of an application or a user interface that may be you know layered on top of the smart contract integrations with other features other tools and technology in the enterprise and really having an end-to-end support plan across that entire piece is again extremely important so that if and when issues do occur there is the ability to go and very quickly find what the problem is and resolve it effectively across that ecosystem so i do want to kind of share a little bit about vmware blockchain and what are we doing to to help bring um our enterprise customers products to production and there are some unique things that i'll just talk about in the context of our product as well which add to a few things to take into consideration um as you go to production so first very briefly um you know a little bit about our um our approach to blockchain we are we vmware have a lot of innovative technology so we have really built this from the ground up there is unique research that has come out of our uh our labs that unique research has been productized and that is what we are bringing to market so this is you know from the ground up built by us it isn't a borrowed open source code and so on right um key aspects that i talked about of about efficiency cryptography performance fast transaction finality granular privacy with with with physical data segregation those are all some of our key innovations um that we have brought that are very um you know core to enterprise use cases we have taken that innovative technology and packaged it into an into a platform with the perspective here really that we want to abstract away all the complexity we don't think everyone needs to worry about all the blockchain technology and all the complexity we want our customers to be able to focus on the um on the applications they're developing on the smart contracts that's really where um they that those are the pieces that are core to their businesses and if we can take off the complexity and the burden of them for operating the infrastructure underneath that then really becomes win-win situation so we've designed our platform to be that kind of an extensible and simple platform to use and then of course what we really are also very aware of is that extreme robustness requirement no one wants to be in the news um for for something going down or something not working and you know these end-to-end testing and the test harnesses we have resiliency in in the case of extreme disasters um resiliency to malicious threat that's a very key piece of what we do and you know performance and um under sustained stress right so even though there are situations with stress the performance continues to be maintained and overlaying on top of that our enterprise grade support which is very natural to vmware to do given the extent to which we support our global customers already so really we kind of take this into consideration as we build our product to stay very true to the blockchain primitives and i will kind of just move on to this layered architecture view that foundational layer really is where we have our unique um blockchain specific technology which is at its at its core we have a byzantine fault tolerant malicious fault tolerant consensus mechanisms um with finality fault tolerance reconfiguration upgrades all in that core blockchain platform along with our granular privacy as part of it we then create what we call virtual execution environments so that you can run multiple smart contracts and not just multiple smart contracts you can run them in multiple smart contract languages so we have support for daimle in production for ethereum um and you know we also then are can extend our platform to have various other smart contract languages supported um we we take this and we really at the end of the day we create a very standard interface on the top so the demo ledger api or the ethereum json rpc api are very standard for what the industry can expect so that the smart contracts developed in those languages can be very easily ported over to the platform or all the tools and technology that that exists for developers and developer environments can be leveraged and then at the same time you can run those applications if we take an ethereum exam example things like the efficiency the transaction finality performance and throughput things like privacy and also things like um i mentioned several of them i'm missing one of the key points yes the gas price volatility right those are kinds of things that make it hard for enterprises to run their business and by doing those things on a platform that enables you to get around um those kinds of issues is a good consideration to be aware of right so i just wanted you to be aware of that and then you know the point i made about this end-to-end support that's really again the point to make here that we want to take care of this entire piece from a support perspective so that our customers can then work primarily on the applications focus on bringing those applications the features and functionality that they need to provide to the ecosystem all right i i want to share a little bit here now about how that architecture that i shared in the previous slide is actually deployed um in in in real life right this is what it would look like you'll see on the right hand side we have what we call our replica network this is the byzantine fault tolerant replica network um that is malicious fault tolerant you'll see that there are multiple nodes um in the replica network these these multiple nodes all look alike so it's a simple um from a deployment perspective of several replica nodes that all look alike and then we have client nodes which really help uh help the application help the participants to connect in into the replica network and it's just those two types of nodes it's a very simple uh architecture the the replica network doesn't have a special database layer or doesn't have special um ordering service or notarizing service or anything like that all that all that functionality is natively in the replica nodes itself so it's you know it is a it is a seamless experience to to manage and monitor across all of them and there are no extra middleware functionality that the customer needs to worry about so this not only makes it simple to deploy and manage and so on it also allows for scalability to be done in various ways as more clients are added just the client nodes can be increased and and the network can be reconfigured for additional client nodes as the network wants to scale in terms of where where clients want to uh operate replica nodes as well the repliconect network can be scaled independently so it's both kind of the the ease of use the ease of support the ease of deployment as well as the scalability that that become simple given the simple architecture i did want to share this um point with with everyone as well because it's a bit uh unique and it's also very interesting as our clients are bringing solutions to production there are multiple ways in which the network itself can be operated so there are three ways here we that i'm showing on the left hand side you'll see that there's a single operator operating you know the core replica network as well as the client nodes and then there are options as you go to the right where the clients can operate then their own client node or they may even choose to operate one of the replica nodes depending how they want to participate and to the extent they are they're willing to take on some of the operational burden but when you think about going to production and you think about trying to reduce the variables and increase the rate of going to production we're finding that the left-hand side is typically where is a good place to start because there the market participants can benefit from the blockchain uh primitives can be on the same um neutralized infrastructure have a single shared source of truth have a single shared automated workflow without necessarily having to take on the operational burden that becomes a quite a good win-win situation where um you know um a trusted third party um sometimes even you know um a jv or or so on can be created so that a centralized operator exists and the entire ecosystem can benefit from it and then moving on to then decentralize it further and further um as the journey goes forward and again that's that's one thing that our architecture is able to provide given the fact that we have both this um the dlt node as well as the client nodes involved and it also uh brings about the way for doing a lot of privacy with with data segregation so that each client node can have its own um um subset of the data and we can do fine granular data segregation between the clients so that each client has only the data that it has is privy to and does not need um the view of all the data on the network i will just um um have one final slide before if there are any questions i'm happy to take them if i have if we still have some time but you know there's this ease of day two operations at the end of the day when you go to production of course getting to production is always an interesting um uh milestone but then this is the long-term management and operations of the of the product in production and the day two operations aspects of monitoring and logging um choice of consumption so that you can put it either on premise or in the cloud the capacity management and upgrades and support that i already mentioned are all very key to how we we deliver successful outcomes for our customers and i just wanted to share this this brief um overview with all of you and hopefully share some familiarity about the work we're doing at vmware blockchain um there's a lot to share more and we'd love for you to reach out to us here are some ways you can reach out either via email through ask vmware blockchain at vmware we also have some locations at twitter you can dm us um and or go to our website um and really appreciate being here ebc and i will go back to victoria thank you yes sorry sorry for interrupting you yeah we are over time tania thank you so much for a lovely keynote er albert has answered a few questions i've also forwarded those questions to to the chat box thank you so much tanya we now have to continue with our last panel of the day the future of crypto exchanges thank you tanya and everyone see you in a bit thank you so much bye-bye bye [Music]