Blockchain ！ The Future of Cloud Computing and HPC
Our computers, laptops, mobile devices, and in the future even our connected cars stay idle for most of the day. Set that against the backdrop that we have a near insatiable appetite for computing power, one that’s increasing all the time, and you have an intriguing opportunity – on the one hand a huge amount of computing power standing idle and on the other, insatiable demand. It’s the old laws of market forces – supply and demand.
Increasingly more entrepreneurs, including John McAfee, and Brian Behlendorf the internet pioneer behind Apache and Hyperledger, are convinced that the future is distributed, and they believe that blockchain, the distributed ledger technology whose potential knows no bounds, is the key to unlocking it all, and making it happen.
Using blockchain they’ve already demonstrated new ways to create the worlds largest supercomputer, disintermediate the giant Infrastructure as a Service (IaaS) public cloud providers such as AWS, Google, IBM and Microsoft, and accelerate the adoption of what’s called Edge, or “Fog” computing. And, as if all of that wasn’t good enough, blockchain has a built in system that will allow people who contribute idle compute to the network to get paid in real time.
I don’t know about you, but that makes me sit and think “What if” for a while, and maybe we’re onto something interesting – something that shouldn’t be ignored.
“There is a growing demand for computing power from industries and scientific communities alike to run large applications and process huge volumes of data,” says Gilles Fedak, co-founder of iEx.ec, a distributed cloud computing platform.
Fedak names several domains, such as product simulation, deep learning and 3D rendering, where demand for expensive computing resources and High-Performance Computing (HPC) is rising.
“The biggest challenge for supercomputing is the demand to compress time,” says Jerry Cuomo, vice president of Blockchain for Business at IBM, “business processes must now be completed at a significantly faster pace than before. The result is that the demand for computing power is increasing exponentially.”
Meanwhile David Sønstebø, founder of IOTA Token, a distributed ledger for IoT, also underlines the need to achieve real time computation and overcome the lag caused by current cloud based models.
“The biggest problem for computation overall is that the devices generating data are not located close by to the data centers that perform the analytics,” he says, “and a distributed network of computers managed by blockchain and smart contracts would help create a shared economy where anyone with a computer can borrow idle computing power and make a side income.”
The peer-to-peer nature of the blockchain and distributed ledgers will also help move computation closer to where the data is being generated, and avoid bottleneck round-trips to cloud servers.
Byrne suggests that while not being a computation platform itself, the blockchain can potentially create “a marketplace application that attacks the specific problem of linking buyers and sellers of compute time and allowing them to pay themselves in cryptocurrency without needing an intermediary like AWS.”
IOTA’s Sønstebø further elaborates on Byrne’s point and says that distributed ledgers “shine in renting out computation in the fog, at the edge of the network.”
IOTA has developed a distributed ledger based on Tangle, a scalable design that gets rid of the blocks and introduces a Directed Acyclic Graph (DAG), a format that reduces transaction times and removes fees, which according to Sønstebø creates the backbone for an outsourced distributed on-demand computational trade model for M2M environments.
Meanwhile Golem, who recently raised $8.6 million in Initial Coin Offering (ICO), to create a “global supercomputer” has based their peer-to-peer computation sharing platform on the Ethereum blockchain, which it dubs ‘Airbnb for computers.’ Golem lets application owners and individual users rent the computing resources of other users and pay for it directly in cryptocurrency. According to Eddy Azar, Growth Hacker at Golem, the platform has the potential to “reduce costs and increase speed in domains such as scientific research, machine learning and graphics rendering, while making it possible for anyone with an average or better computer to share resources and make a side income.”
After submitting tasks to the Golem network, requestors are matched with providers based on prices, reputations and machine performance registered on the ledger. Resources are then sent to the provider for processing and are sent back after the task is completed, ad the provider is paid if the results pass verification tests. A user’s reputation is built-up based on their success in completing tasks and making payments.
Developers can use Golem’s open-source code and task API to create program that uses the network and put it on Golem’s application registry, which Azar likens to an app store, and make it available for others to use and, if the developer so chooses, pay for.
iEx.ec is another distributed computing platform, which uses the Ethereum blockchain to create a market network for applications, data, and computing resources – including HPC.
“It means that everyone will be able to offer their computing resources through smart contract deployed on the blockchain,” says Julien Béranger, the company’s Community Outreach Officer. In iEx.ec’s case the platform uses Desktop Grid or Volunteer Computing to collect underutilised computing resources to execute very large parallel applications at a fraction of the cost of a traditional supercomputer. This is the model used in distributed computing platforms like University of California Berkeley’s SETI@Home, Stanford University’s Folding@home and disributed.net.
The iEx.ec team hopes the combination will provide inexpensive, scalable and on-demand access CPU, GPU, data sets, storage and other resources.
Blockchain makes a big difference, iEx.ec’s Fedak believes.
“Because the blockchain allows for a decentralized infrastructure, it can bring the data closer to their producers and consumers,” he says, “whereas with centralized cloud computing, data-centers tend to be located in remote areas.”
The demand for computation will continue to grow as we move forward. Whether cloud servers will scale up to meet the requirements in resources, costs and speed is yet to be seen. In the meantime, the blockchain proposes a number of intriguing new alternatives that can open up new possibilities and succeed where previous technologies have failed.
By Matthew Griffin