# iExec: Fully Distributed Cloud Thanks to the Ethereum Blockchain - Dr. Gilles Fedak

- Speakers: Gilles Fedak
- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2017-10-09
- Duration: 12:47
- Topics: blockchain, smart contracts, distributed ledger
- Watch: https://streameth.org/watch/yt-aCpv3qLZJWU
- YouTube: https://www.youtube.com/watch?v=aCpv3qLZJWU

## Description

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on IPFS and more.
https://archive.devcon.org/archive/watch/2/fully-distributed-cloud-thanks-to-the-ethereum-blockchain

iExec provides the SMI / SME and individuals a scalable, secure and easy access to the application, the data-sets and the computing resources they need. iExec leverages a set of research technologies that have been developed at INRIA and CNRS in the field of Desktop Grid computing (aka Volunteer Computing). The principle is to collect the computer resources underutilized on the Internet to execute very large parallel applications at the fraction of the cost of a traditional supercomputer. Since the early 2000’s we have pushed this idea to its extreme limit by proposing many software and algorithms in the area of large scale data processing, data management, parallel computing, security and dependability, QoS,… iExec relies on XtremWeb-HEP, a mature, solid, and open-source Desktop Grid software which implements all the needed features : fault-tolerance, multi-applications, multi-users, hybrid public/private infrastructure, deployment of virtual images, data management, security and accountability, interoperability with eScience infrastructure, and many more. In this talk, we’ll show how Ethereum can be used to organize decentralised infrastructures and market places, where everyone will be able to rent its computing resources (CPU, storage, GPU, …), where the innovative SMEs which design Big Data and HPC applications will be able to sell them online immediately with the needed resources to run them, and where highly valued data-sets will be rentable with a fine-grain business model. We’ll present some early results that illustrate the interaction between the Ethereum blockchain and regular distributed systems (e.g task scheduler, data management, etc..) Moreover, because the blockchain provides distributed consensus and resiliency while being fully distributed, we think that it’s going to change drastically the way we design distributed systems and applications. We’ll draw some perspectives on blockchain-based infrastructure management, and present the related research directions.

Speaker(s): Gilles Fedak
Skill level: Advanced
Track: Society and Systems
Keywords: fully, distributed, cloud, blockchain, infrastructure, dapps, security, resilience, miners, computing, power, iexec, offchain, cpu, servers, HEP, scheduling, resource, management, autonomous, sidechain, energy

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon 2 was held in Shanghai, China on Sep 19 - 21, 2016.
Devcon is organized and presented by the Ethereum Foundation, with the support of our sponsors. To find out more, please visit https://ethereum.foundation/

## Transcript

so thanks everyone my name is Jeff attack I'm a researcher at inria which is the French national research institute for computer science and my research background is in parallel and distributed computing with a focus on how to build distributed computing infrastructure under using Internet resources the work I'm going to present today is actually a joint work with Hyrule hair from the Chinese Academy of Science I don't know if I will you are here yes he also is the guy here and this talk today is about how we can build a blockchain but fully distributed and cloud and what would like to my goal today is to give you a perspective from the infrastructure point of view which i think is very important so as you know if Iran allows you to develop this trivial application distributed application and systems that run on the blockchain and the blockchain gives this application of some very nice properties applications are going to be autonomous they are going to be a resilient secure and then there's also these distributed consensus so that's a great thing that's great promises and particular sometimes we are talking about a supercomputer but actually what's happening when you try to move your existing application to the blockchain is that you face the fact that in fact you are offered very limited computing capacities offense and storage is quite expensive it spare you have a slower execution very high latency if you need communication through the transaction and so that's really a limitation and so and so the question he is a hooker and with I mean I live yet this and that's especially an issue if you consider the fact that there is a huge computing power which is given by all these are the miners so there's really a gap and the objective of this work is so to try to give all this computing power back to the application that needs it so the objective of i exac is to provide blockchain by distributed application access to ocean computing resources when they need it so in terms of computing resources were thinking about CPU we are thinking about data access application in particular focusing on compute intensive application services and so forth and to do this of course we can take advantage of the blockchain and try to organize a global market for computing resources so think it as a kind of Airbnb for servers where everyone would be able to put its own servers and use someone else evers and so that's still a cloud in the sense that you access through the blockchain to the computing resources on demand and as a payoff as a pay-as-you-go basis and so this idea of building a distributed cloud is actually very timely at the moment what's happening with the cloud computing is that very centralized on data servers and big data centers and I don't know what the situation your respective countries but in France for instance if you want to set up a new data center in Paris area the answer is simply no not enough room not enough power forget it and so this little situation where you have all those data centers that go in very remote area you know to have either free energy or free cooling like cheaper energy or free cooling like Iceland in Europe or tiburtina in China for instance and so distributed cloud computing is also about relocating those data center closer to the data producer and consumer and just to give you an idea of you know how we could read some next-gen data center distributed data center is that some picture coming from people who are working with so on the left you have the Reuters University they set up the data center on the roof top of their building with solar panel and this is fully you know low-power processor batteries and this is fully energy autonomous under right oh sorry on my left and so on your right this is Colonel computing the blackboard that you see here is both a server a server and a heater so this is actually the CPU you notice the CPU energy which is warming when doing 3d rendering which is warming your apartment during winter winter time so we can distribute the data sometimes feasible I'm always coming with edge for computing of course the goal of I exactly tooth is to make those not be able to to to add those kind of it right so to be run on the blockchain so how to do that actually the technology to build a dead distributed cloud is already there since many times we just call that desktop grid computing it's the idea of using PCs and the internet when they are either and it's very mature technology so for instance during those everything to do the distributive cloud you have security virtualization can chip VM images you can even do quality of services you know high level things so for instance in Europe we had the European Union funding the European disturb green infrastructure we had about 11 different sites in Hungary France Denmark etc totally not two hundred thousand computing nodes and this was available you know to the traditional science infrastructure such as the LHC so we were able to transparently execu many applications coming from finance biomedical research mathematics high-energy physics etc so in my research team we used to push this idea in every possible direction with the parallel computing and internet Map Reduce computing and internet quality of services for execution while machine da volatility setter etcetera so for this work what is interesting for us these are two pieces of middleware basically extreme web HTP which is if you want the production version one of these ideas it's done by your Oleg ludajenski at the in2 preframe it does task management and the other one is bid you that those data management so the way we are working at the moment mostly you know experiment and proof-of-concept see what's going on so we have the regular stack if you want the application that has no specialty blockchain pass the resource management middleware and and of course the executing the actual machines and what we do is that we put a film in between I would try to see what's going on what is difficult or challenging or what is easy that's all that's basically the result so on the left part you have what is super easy the goes we'll show it in green and on the right part light right path it's what is red and what is really challenging so results publication this is basically you have you know result description language or very well standardized and you basically translate this into some sort of act contract and put it on the blockchain so fairly easy resource provisioning just add the flag that gives the status of the resources and then it gets more tricky so for instance match matchmaking is you want to be able to say this with this application that requires this amount of memory can run on those resources so you have to brought some holder blockchain and then things get real tricky with scheduling so simple scheduling so scheduling it you have a list of tasks you have a list of machines and you want to build the execution plan so things get really tricky when you need you know advanced scheduling advanced scheduling with multi criteria so things like I'm ready to I want a fastest execution possible and I'm ready to pay for that it's getting really a computer tension memory intensive so no use case we're also working with an application it's called if a this is a framework for doing Finance analysis and the interesting things here that it does a lot of machine learning so a lot of machine learning it's both data intensive and compute intensive so here of course I'm not going to make it on the the blockchain and you need to go offline I've seen so and the good thing is that once if a service is ready their customers will be able to rent it directly on the blockchain because this application is going to be autonomous it will fetch itself the resources in need on the blockchain to execute it option okay so our test bed for doing this it's the grid 5000 is the French interaction show dedicated for researching in distributed system so we are very lucky to have this at all disposal it's actually a distributed cloud it's nine sites and we have 100 about of course 1,000 nodes 8000 course and yes and the and so 3 grid to do you know scalability tests and this kind of things and we have a dedicated 10 Giga Network and so the last I mean the bottom in or use case it's we want to use the steam LG servers to run this so so steam adji what they are doing it's a this is actually that this server is a furnace so it pre-warmed the the water before distributed this in the building and we hope to to do demo 18 know in November at the supercomputing a conference and that will be for the first time in the history of blockchain a smart contract with wood side effect is to warm a swimming pool so amazing this is true this is true this is lab it ok in Paris so bad on those earlier experiment that's a sketch of what the future is going to be for us most certainly so from what I am what I think is that we are going to have something like a side chain because we need know some sort of a new consensus for what's going of shame that we call that proof of contribution important thing is that there are many information that you need to make it run but you know that are not necessarily relevant with respect to the to the to the provisioning contract if you so for instance everything that is about tasks you sure this is meaningful when you want to establish the fact that the computation went right but it's not really meaningful with respect to the contribute to the contract that provision those resources and as for the the execution so I want to I think we need some dedicated section for this we also have some very specific workload in in perilous and parallel computing you often have things that arrives really in bursts so so we have to to be able to optimize for this specific workload and and last thing is that the console fees can be very different for parallel computing you know in some case in parallel computing that's totally okay to have a fraction of your application that gives you wrong result it just depends on the application so we must be very flexible with respect to this okay so just to conclude distributing the cloud I mean it's also give us some new opportunities that's very important so as I said I'm repeating myself it hats which are matters and so as we can set up the rule that's our own game so we can do whatever we want so why not for instance think of an infrastructure that would be energy positive no that would produce more energy that in consumes I mean a charger up to us to decide this okay so just to find some some dude and one showing the blockchain lab Dacca had the French team who has been a very supportive with us thank you very much