Over the past year, I have become heavily involved in the world of blockchain.
In addition to advising a number of blockchain companies on their overall communications and messaging efforts, such as ShipChain, RedPen, MagnaChain, and more, over 25% of our clients at Digital Press are founders and/or board members of blockchain companies, looking to separate themselves from the competition.
What’s so intriguing about the blockchain space is the fact that with this new technology, some of the world’s most glaring issues start to move more and more into focus.
Here’s a mind-blowing stat:
Researchers estimate that less than 0.5% of all collected data is ever analyzed.
As powerful as big data has become in today’s day and age, we are still a far cry from fully leveraging the mass amounts of information being collected.
Many companies lack the infrastructure, personnel, and experience needed to efficiently sort through these massive data sets — and instead export these tasks to a select handful of technology companies which possess the necessary resources and power. As a result, businesses like like Amazon Web Services and Microsoft Azure currently dominate the market, charging premium prices based on the computing power required by consumers. And conversely, many small businesses and start-ups struggle to afford these services, limiting their ability to extract insights from their company’s data.
Combined, these challenges and potential market monopolies create a unique business opportunity for innovations in the cloud computing space.
For those that don’t know, cloud computing refers to the practice of exporting “problems” to remote servers when analyzing data to minimize the energy burden on a company’s local network.
Generally, there are four types of cloud offerings:
1. The private cloud is a service operated solely for a given organization, providing companies with the ability to customize their digital infrastructure and ensure data confidentiality.
2. The public cloud is hosted by service providers and is more commonly used by the general public.
3. The community cloud is a shared source of infrastructure between multiple organizations with a shared purpose or function.
4. The on-demand model provides companies with access to a shared pool of configurable computing resources.
There is also what’s called “the hybrid cloud” model that combines two platforms (e.g. public and private) through standardized technology.
Over the past 5 years especially, cloud computing has propelled a major shift in productivity — in the workplace, between remote teams, etc. — and fundamentally changed the way people use and access digital applications (e.g. file sharing through public clouds like Google Drive or Microsoft OneDrive).
However, cloud computing remains an intensive process which most businesses lack the capacity to perform in-house.
One potential approach to reduce the resource burden of cloud computing would be splitting up processing activities such that multiple computers can work on a task simultaneously.
This concept is known as “distributed cloud computing,” which refers to technologies that connect the computing power of many computers on a decentralized network to reduce resource use and maximize computational power.
Suddenly, computers connected within a network can communicate by passing messages about which portion of the puzzle they’re working on, and whether they’ve solved it yet. Remote Procedure Calls (RPC) or Remote Method Invocations (RMI) then distribute a computational task among available computers so they can work in parallel. However, while the untapped potential of distributed computing is easily appreciated, distributed computing systems in their current state don’t independently have the high availability or the failure resistance that would be needed to run a site like Facebook, that uses about 50 million operations per second.
Together, however, distributed cloud computing proposes cloud infrastructure supplemented by the distributed computers in multiple geographic locations.
The increasing popularity of blockchain decentralized applications, or dApps, are subsequently placing a growing computational burden on the blockchain ecosystem. As dApps become more complex and computationally expensive, the ecosystem needs a decentralized marketplace for cloud resources.
Companies actively rising to providing distributed cloud computing services on the blockchain to the market include Ankr, Golem, SONM, DFinity, IExec, Perlin, and Hypernet. Today, most services are Ethereum based. They all provide cloud computing in some form, often through smart contracts on a blockchain.
Golem and SONM are similar, but target different niches of the market. Both provide computing by using third party cloud computing containers like Docker, to manage the different needs of multiple clients. Users have a single source of income for cloud computing based on the Ethereum platform. IExec goes a bit further, using an existing grid computing platform to provide computing units rather than outsourcing to a third party. IExec is currently Ethereum-based, but aims to be platform agnostic in the future. Dfinity gets a random selection of the miners involved in computation, but remains limited by the smart contract because of where the work is executed.
Perlin is more privacy-centric, using virtualized computing containers much like Golem. This early stage technology proposes a more isolated configurable infrastructure to enforce greater security. And finally, Hypernet is developing a parallel computing API that will improve the communication abilities between computing nodes. Hypernet’s execution is tied to the smart contracts, and the team plans to use containerized computing environments; however, these platforms still carry but these are also subject to security threats and data privacy concerns.
Where most of their competitors focus on improving transactions per second (TPS), Ankr is focused on providing a more viable distributed computing platform.
Ankr proposes innovations at the consensus level. To this end, Ankr finds that the time for Proof of Useful Work is now. Bitcoin chose to rely on mathematically arduous hash calculations specifically in order to mitigate misbehavior. The game theory of this argument changes when the adoption of Trusted Execution Environments (TEE) is considered. TEEs guarantee execution will happen exactly as expected. Given that the upgrade time period of a computer is around 2–3 years, and TEEs were released about 3 years ago, 2018 is a good time to watch the TEE adoption rate, and see how if affects the underlying blockchain technology. With TEE, a node owner won’t be able to view the data and jeopardize operations. Furthermore, by integrating Ankr into the existing BOINC project network, they are immediately elevated to around 300K+ nodes worth of computing power.
Ankr’s Proof of Useful Work is not constrained by the limitations of a smart contract like Dfinity, particularly because it is executed directly in the CPU. By using the miners for cloud computing rather than a third party, users of Ankr are provided with three sources of income: mining, transactions and cloud computing. With many other platforms, the computation fee is the only reward for someone filling a computation request, meaning that the baseline computation fee must at least cover electricity costs to break even. Using the Ankr platform however, miners collect both the computation fee as well as a mining reward, which enables them to accept smaller computation fees. Thus, Ankr is able to provide their clientele with lower computation costs for the same amount of work done. Ankr also has higher computing efficiency, because all miners are participating, rather than Dfinity’s random selection.
While distributed cloud computing holds tremendous market potential, the technology must continue to grow and adapt to demonstrate a capacity for scalability.
Technologies like Ankr, which build off of large existing networks and represent fundamental transformations to the increasingly outdated consensus mechanisms of cryptocurrencies highlight the exciting potential for distributed cloud computing to support business applications through the blockchain.