Cowritten by Alex Price, Founding Partner of Qu Capital
Decentralized Cloud Computing. I’m not sure there’s ever been a less sexy combination of buzzwords. But those three words should get you excited — very excited. And if Decentralized Cloud Computing excites you, then iExec should have you ecstatic.
To understand the significance of iExec, we need to understand the industry as it stands today. The cloud provides access to processing power, storage, and applications to companies without the expensive equipment and more importantly, without the difficulty of fixing and optimizing hardware.
Centralized cloud computing is a multibillion dollar industry — $206 billion as of 2016 and by some estimates, it could break $400 billion by 2020. It’s clearly a high demand product. And it’s not difficult to understand why. Cloud services replace massive on location data and processing centers for companies. They allow companies to forgo expensive hardware investments and focus on running their business. To avoid paying for large IT departments, manage hardware upkeep, and pay for physical space, companies outsource to cloud companies for their processing, storage, and bandwith needs. Simple and easy. With cloud computing, companies also only need to pay for what they use meaning there is no excess spending. Netflix today is entirely powered by cloud computing provided by Amazon.
Some of the biggest companies in the world provide cloud computing services. Companies like Amazon Web Service (AWS), Microsoft Azure, Google Cloud Platform, IBM Cloud, and Alibaba Cloud account for hundreds of billions of dollars in revenue per year. These companies have enormous data centers throughout the world, each with between 50,000 to 80,000 servers.
And the demand for these services is growing exponentially every year.
TLDR: Instead of purchasing and maintaining large in-house servers, companies can rent computing services and storage space from cloud companies.
Decentralized cloud computing in its simplest form is doing what AWS and Google are doing, but in a decentralized manner. Instead of having massive centralized data and hardware centers, the computation comes from individual computers all across the network. When your computer is off and you’re getting some much needed shut eye, you help the Request Network smart audit your father-in-law.
Some companies like Sia and Filecoin are providing decentralized storage solutions. With these, you can rent out your storage space to other people. But what iExec is developing, is a cloud computing solution. That means that companies can rent the idle compute power (not storage space) from the decentralized network. Now you, savvy investor and critical thinker, may ask: how can iExec compete with Google and Amazon and Microsoft? How can a blockchain startup compete with multibillion dollar corporations that effectively own our digital identities? How can they provide a cheaper and more effective solution than these global superpowers? The answer: Right now, iExec can’t. But It doesn’t need to compete. iExec can be the AWS for blockchain.
TLDR: Blockchain enables decentralized cloud computing. This enables users with excess computing power (most of us) to rent the power out to those who need it.
iExec is a decentralized application (dApp) running on the Ethereum network. It enables access to decentralized computing power, not only for individuals, but for dApps on blockchains as well. In addition to my research, I interviewed Wassim Bendella, the Blockchain Business Developer for the iExec team. Specifically, I was looking for clarification on some of the technology and confirmation that my understanding was in fact, correct.
Ethereum is revolutionizing the Internet by providing programmable trust, in the form of smart contracts. These smart contracts act as trustless middle men, ensuring all parties stay honest, without the added cost and delay of lawyers and contracts. With over 800 decentralized applications, Ethereum provides a secure, decentralized, and trustless development platform for the new Internet of Value. These new dApps utilize Ethereum for their smart contract and blockchain needs, storing the terms and state of the contracts on the blockchain. Ethereum’s virtual machine is responsible for the execution of smart contracts. Essentially, in layman’s terms, virtual machines are machines that execute smart contracts and run inside the computers of miners or other validators. The VM is the environment that the smart contract runs in. And Ethereum’s VM is excellent for simple transactions and business logic — things like managing the rules regarding the creation and storage of information. And although Ethereum is Turing-complete and theoretically capable of performing any computational feat given infinite time, the VM struggles when performing rigorous computations efficiently and at reasonable cost.
Good luck with that car chase…
That’s not what Ethereum was made for. Running intensive computations on Ethereum would be like trying to run Grand Theft Auto 5 on Apple’s 1987 Macintosh II.
Today, this isn’t really an issue because most dApps don’t have a working product. The dApps only utilize Ethereum for their ICO sale and for simple transactions with their tokens — Ethereum is well suited for that. But once all these dApps start releasing actual products, with actual computational demands, Ethereum not only will not be able to keep up, but it will become prohibitively expensive. Since Ethereum charges ether (gas) to run the network, intensive computations would cost a fortune. Imagine when Ethereum’s computational demands increase by 10x, or 100x: the network will be unusable.
TLDR: Ethereum is great for simple computational processes like transactions. But the dApps on Ethereum need to do more. They need more power than the Ethereum virtual machine can provide.
The solution to this is offchain computing. dApps need to be able to do their heavy computing offchain (meaning not on the Ethereum blockchain) and then bring the information and results back onto the Ethereum blockchain for verification. iExec enables just this. dApps can use iExec’s decentralized cloud computing services to safely, simply, and cheaply, execute their computation far from the inefficient and incapable Ethereum VM.
Maybe the AI could finally beat my dad for me…
An interesting use case that demonstrates this relationship was an AI chessbot developed by Stockfish. The actual chess moves are a simple recording of information — Knight to C6 — easily accomplished by the Ethereum VM. But the actual AI decision of where to move is exceptionally complex. The amount of information that the bot needs to process is enormous — strategy, combination of moves, anticipation of opponent’s moves, etc. — this is impossible for the VM. So iExec provides the computational power for the AI off-chain, and then transfers the move results back to the Ethereum blockchain where they stay, immutable and secure; all this happens fast: well within the time for one Ethereum block.
The use cases are endless. Flixxo is a dApp that envisions itself as the decentralized YouTube. Because of all the encoding and decoding of videos on the blockchain, it requires an enormous amount of computer power. “iExec will allow the platform to do [this] by providing cloud resources,” says Bendella.
Request is another partner. Request plans to offer automated auditing of financial records. This is also an extremely intensive computational process. iExec again provides the offchain solution for this computation — the results of the audit can then be easily transferred to the Ethereum blockchain.
The blockchain economy will need a service like iExec. In 2018, as dApps begin releasing their platforms, it will become increasingly clear that Ethereum will be ineffective for all that computation. With AI, smart cities, VR, and IoT all utilizing advanced smart contracts, there will need to be an offchain computing solution. iExec is that solution.
I also don’t want to mislead. Users and businesses will also be able to utilize the iExec services for their own projects and uses. Need more power for your fantasy football algorithms? iExec can do that. Want to play Fifa with 4k settings? iExec is your tool.
TLDR: iExec processes the computations away from the blockchain and then transfers the results back to the blockchain. dApps can use this to perform complex computations like rendering, uploading, and all the requirements for AI, IoT, and VR, all without slowing down the Ethereum network.
iExec didn’t create its foundational technology, it only invented a way of combining the technology with the blockchain. iExec uses XtremWeb-HEP, an advanced software developed for cloud computing. This software is a peer reviewed, open-source technology produced with the help of the iExec team during their time in cloud computing research. The technology essentially helps solve many of the issues associated with a service like this including: fault-tolerance, supporting multi-applications, multi-users, and hybrid public/private infrastructure, and ensuring security and accountability. Which is just a long way of saying: it provides solutions for all the considerations of a project like this.
iExec already has a working platform. Their Software Development Kit (SDK) was a part of their V1 released during the beginning of November. The SDK allows dApps to utilize iExec’s computing power. This means that today, dApps can already perform offchain computations using iExec.
By building on XtremWeb-HEP technology, iExec has constructed a trifecta of cloud computing, its Triad of Technology.
iExec is broken down into three distinct platforms that all interact to bring decentralized computing to both individuals and dApps.
The dApp store released on December 20th. The dApp store is the equivalent of Apple’s App Store where all the dApps utilizing iExec’s services are featured. Users can then interact with all those applications and pay in iExec’s native token: RLC. As Bendella said, “[The dApp Store] allows us to link the whole dApp ecosystem powered by us. Users can browse dApps, use them, and pay for them. Application providers can list their dApps and earn money if they decide to monetize them.”
Currently, the dApp store is only an interface where users can view dApps built on iExec. However, they plan to expand it beyond simply cosmetics very soon. This will allow users to actually engage with the dApps from the store.
The iExec dApp store
TLDR: The dApp store provides an organized and visual platform where users can see and interact with the dApps using the iExec service.
The marketplace is the bread and butter of the iExec platform and will be released during the next update in Q1 of 2018. The marketplace essentially enables the buying and selling of computing resources. Users — you and me — can sell our excess computing power in the marketplace. dApps can then purchase and utilize those resources. Right now, even with the launch of SDK, dApps are only utilizing computing resources provided by servers maintained by iExec. This centralization is a short term limitation for iExec as the computational power is currently limited. However, once the marketplace is operational, it will truly be decentralized as many users will collectively provide the necessary resources for applications.
This decentralization will enable cheaper costs and superior scaling. The marketplace is based on free market economics. The more users selling their idle computer power, the more competition there will be, and thus, the cheaper the fees. iExec is only limited in how much computing power they can provide by the number of users selling those resources. As long as there is a growing base of people selling resources in the marketplace, the sky is the limit in terms of iExec’s scalability.
“Right now,” says Bendella, “iExec provides computing power for free. In May 2018, computing providers will be able to monetize their own resources on our marketplace.”
The marketplace will also feature a reputation system that values a user’s reliability. With this, smart contracts will be able to specify price, speed, and reputation requirements for their computation needs. iExec has a complex matchmaking system that enables this service. For example, I might be willing to sacrifice reputation and speed for a cheaper computation price. Other users may be willing to pay more in exchange for a faster and more reliable computation. All this is possible with the iExec marketplace.
Users pay in RLC but the iExec network charges no fee. This allows the marketplace to provide a safer, cheaper, and more reliable cloud computing solution to individuals, companies, and applications alike, especially as technology continues to scale.
iExec can also provide means for massive mining pools and data centers to rent their computational power in the marketplace. There is a massive amount of available computational power in these pools.
TLDR: iExec will have a marketplace where users can sell their excess computing resources to those who need more resources.
In our world of Big Data, companies have enormous amounts of data and no idea how to utilize it. This data could be anything from stock market or medical data to enormous quantities of consumer data. There are plenty of applications and services which could utilize that data and are willing to pay for it. This data marketplace enables that matchmaking. Companies that have excess data can sell the data to any application that wants to purchase it. All of the transactions are paid in RLC. Any data legally accumulated by a company can be sold.
“Today in the world of Big Data, we have massive datasets waiting to be turned into value. Facebook and Google do it well, but I think this can be done by anyone. Therefore, we are building a marketplace to connect those that have data to those that don’t have it but would like to leverage it. Applications running on iExec will be able to make use of an ocean of data at their disposal.” — Wassim Bendella
It is important to recognize that this service is a ways away. The dApp store is releasing soon and in early 2018, the marketplace will be operational. However, the data marketplace will not be released in the near future.
TLDR: iExec will have a marketplace where companies with large stores of lawfully acquired data can sell it to businesses and applications that can make use of said data.
Imagine I have a fantasy football application — OneMoreInch — that provides advanced predictions for fantasy football users and scorekeeping for head-to-head games. The application’s prediction algorithms use the iExec marketplace to find users who are selling their computing power. The complex algorithms are intensive and require huge inputs of computing power. With the extra power provided by iExec, OneMoreInch can smoothly run the algorithms. In addition, with the future creation of the data marketplace, OneMoreInch can aggregate player and fantasy data to effectively compute the predictions. All the scores and results are then transferred onto the Ethereum blockchain.
If a user wants to use OneMoreInch, all they must do is go to the dApp store and pay in RLC to interact with the service.
An overview of the iExec triad
One of the main challenges for decentralized cloud computing is how to prove who provided what computational power and compensate them accordingly. iExec has designed a system called Proof of Contribution that certifies resources and credits them to the external providers. This record is maintained on the blockchain.While the exact details of this have yet to be provided, this will hopefully provide a trustworthy system that enables accurate consensus regarding who provided what in the iExec transaction.
While iExec is currently built on Ethereum and all my described use cases revolve around Ethereum, iExec transcends this specific blockchain. iExec can adapt to any other blockchain — NEO, EOS, Cardano — and provide the same off-chain computation services. “We’re focused on Ethereum right now, but we can move our technology to almost any other blockchain to suit some specific need,” said Bendella.
The iExec Token: RLC
All network fees are paid in RLC. Because there is a maximum supply of tokens, as demand for the network increases, so will the price of the token. As more dApps begin utilizing iExec and transacting with RLC, the price will increase.
Total Supply: 87,000,000 tokens
Marketcap: $79 million
Token Distribution:
69% released in ICO
17.2% held by the team and advisors
6.9% held as an emergency contingency fund
6.9% held by iExec for network incentives and development
Currently, only authorized smart contracts will be able to be deployed on the iExec blockchain. That means that there must be a peer-review process that ensures that all smart contracts are legitimate and trustworthy. Not many details regarding this process have been released.
V1 released according to schedule for iExec, bringing their decentralized computing solutions to dApps. It is very impressive that iExec was able to deliver their V1 on-time. It gives me much more confidence in their team and project. With all the difficulties and unforeseen obstacles of developing with a new technology, many projects in this space are unable to stay true to their timelines.
Their dApp store releases tomorrow. Their V2 — the decentralized marketplace — will release in May of 2018. This will enable users to sell their computing resources to others.
The team is one of the most exciting aspects of iExec. I like to invest in teams. Ultimately with projects, no matter how flashy their marketing or how impressive their white-paper, it is the team that is developing the project. The challenges of producing a working product in this industry are immense; it is critical that the project has a strong organization that can effectively respond to those challenges.
iExec’s team is the kind I look for. The two founders of the iExec project, Gilles Fedak and Haiwu He, are both powerful people in the cloud computing industry. Fedak has 14 years of experience in cloud computing research at INRIA, a public research body dedicated to digital science and technology. With a Ph.D in Philosophy and Masters in Computer Science, he has been instrumental in many technical developments and has coauthored about 80 peer reviewed scientific papers, and won two best paper awards.
Haiwu He has an equally impressive job history. He is a professor at the CNIC (Computer Network Information Center), Chinese Academy of Sciences in Beijing and a scholar of the Ministry of Education of China. He also worked as a researcher for INRIA in France and has published about 30 scientific papers. He holds a Ph.D in Computing.
Together, they were instrumental in founding the Desktop Grid Foundation which is an organization that brings together experts in the field of cloud computing.
The key to understand about them is that they are Ph.D holding experts in this field. They understand the industry, the requirements, and the vast considerations required for this project. They are published, accredited, and well respected. They are leaders whose over 20 years of research have been committed to distributed grid computing research. These are the people I want running this company.
However, they are not marketers, and it shows. Their goal is not to advertise their product and their lack of marketing has hurt RLC’s valuation. Their whitepaper is drudgery, and they don’t have a strong social network presence. However, their production has demonstrated their competence in bringing cloud computing to blockchain.
In addition, I expect that their media presence and marketing proficiency will improve dramatically. Their marketing team was put together only two months ago and is already showing promise; they released a website redesign that is far superior to the previous one. The iExec team will also be releasing a revised and updated whitepaper during the first half of January.
TLDR: iExec’s team has years of industry and research experience, has published over 110 academic articles, and helped establish a cloud computing research organization.
Golem is similar to iExec. However, they are currently focusing their marketing on a specific use case: digital rendering. Digitally rendering images and animations is a very GPU intensive process. Golem aims to use decentralized computing to enable extremely fast digital rendering. Ultimately, Golem is initially trying to attract 3D artists, animators, and designers while iExec is creating a cloud marketplace for users and dApps. Potentially, they could even cooperate. While they could one day compete, I think for Golem, their choice in marketing speaks volumes. iExec’s main value lies in providing offchain computation for blockchains which is a different objective than Golem. Golem has first mover advantage with this use case and iExec has the advantage in providing dApps with decentralized computing solutions. I don’t see them as being in competition at this time.
SONM is also working to enable a decentralized super computer. However, I have very little confidence in SONM’s ability to deliver. They are attempting to provide decentralized Fog and Edge Computing (their definitions are well above the scope of this article). Fog and Edge are both enormously complicated and not realistic for initial stage development. Both iExec and Golem plan to expand into that field, but recognize that the complexity demands a serious foundation and already working platform. Any strong engineering approach should focus on slow repetitive development: carefully building each foundational block first to achieve an end goal. The fact that SONM is approaching the Fog and Edge computation from the beginning does not bode well for them.
While iExec may in the future try to provide decentralized cloud storage solutions (essentially renting out computer storage space from users on the network), they are not currently doing so. iExec only deals in decentralized computation. Thus, they’re not in competition with these applications.
TrueBit was intentionally not included. While they may one day compete with iExec, this is likely over a year away. With how rapidly this industry evolves, a year is like a millennium; at that point, there will be a whole new set of demands and considerations. I focused only on their competition now and in the near future.
TLDR: There is no true competition for iExec. Golem is approaching decentralized cloud computing from a different direction and I have serious reservations about SONM.
I have mentioned that I don’t see iExec as needing to compete directly with Amazon, Google, or any of the cloud computing heavy hitters. And in the beginning, due to the their current centralization and the likely slow adoption of their marketplace, iExec won’t be cheaper or more powerful than AWS or Google. However, iExec’s decentralized model offers potential that could exceed any of the major players. If iExec can facilitate a robust free market for cloud services, it’s low cost and high scalability could, in the long run, be unmatched even by Amazon. There are truly few limits to the scale that a decentralized marketplace like iExec could reach.
If there is one aspect of iExec that I hope this article highlighted, it is this: Ethereum cannot support the kind of computational needs that many of its dApps will require. This problem will compound exponentially as more dApps release their working platforms and products. Blockchains need a solution that allows them to compute offchain and bring only the results on chain. iExec is that solution. No one else is doing what iExec is doing. Not Golem, not SONM. The extreme need for offchain computing will be there soon: iExec will be well situated to service it.
Sitting at close to an $80 million marketcap tells me that they are severely undervalued in comparison to their peers: SONM and Golem. Here are the reasons I think iExec’s valuation should be higher. They have:
The reason that it’s so criminally undervalued in my opinion is that people truly don’t understand iExec’s functionality and potential. It honestly took me a long time to understand why decentralized cloud computing is critical to blockchains. It’s not intuitive and certainly not flashy. In addition, their lack of strong marketing has significantly contributed to iExec’s $80 million marketcap. I see iExec’s immediate valuation as being close to Golem’s (a $480 million marketcap) which would represent a 600% increase. Truthfully, I think that is conservative.
It’s also important to understand that iExec doesn’t need to compete with Amazon and Google; and while iExec’s decentralized model will offer better security and energy efficiency it likely can’t compete with the cost efficiency of these global superpowers. But iExec’s purpose isn’t to compete. These massive, centralized cloud companies cannot service the new Internet. Decentralized applications can’t have centralized computing solutions — that defeats the whole purpose of decentralization. iExec is providing what Amazon and Google cannot.
When people recognize that iExec is the future of offchain computing for blockchain, I am confident iExec will finally get the recognition that it deserves.
A big thanks to Wassim Bendella for his comments. And a thanks to Nir Kabessa for his edits.
FOLLOW me on Twitter: @noamlevenson
Disclaimer: Both Alex and I are invested in RLC. This is not investment advice, merely our opinions on the project. Do your own research. We did not receive payment from the iExec team and this is in no way sponsored content.
You can buy RLC from Bittrex.com
Donations and support are greatly appreciated
NEO: AGpjyU9FY1H58tGdhhBxgTxaosxh5uUmvp
ETH: 0x4c35100a0a25b3933ba1c0469b5df8b24035775b
BTC: 1Dmri6J1epQ7MG1THCWvpQAfLAjeSxe1hx
Hold down the clap button if you liked the content! It helps me gain exposure .
Clap 50 times!
I love getting questions or suggestions, so comment away!