Disrupting Computers – Power in Sharing

Making a business out of shared processing power

In the past, if you needed a resource you didn’t own, you would simply have to buy or borrow it. Thanks to the rise of the sharing economy, instead of paying outright, consumers can now rent services or products from their peers, bypassing the costs and complications associated with big businesses. Now, this relatively new economic model is influencing computing power. It might be common to own a computer, but it’s not always easy to access the required amount of power for certain tasks. Fortunately, Polish startup Golem is aiming to address this by creating the decentralised Golem Network, combining users’ computing power to do for computing what Airbnb did for accomodation. In November last year, the company raised $8.6 million from Initial Coin Offering (ICO) sales, and is quickly shaping up to challenge incumbent, centralised systems. So, how do P2P computing networks work, and how disruptive are they?

A computing community
The speed of computer processors is measured in cycles, which represent the time taken to receive a programme instruction and carry it out. The more complicated the computing task, the more cycles are required. Take CGI, for example. One PC alone may take days to render CGI programme instructions, but share the task between multiple computer cycles and it could take seconds. Golem allows users to distribute jobs within its network for less cost than traditional cloud based services, and without a reliance on a single authority. Payments are made via digital Golem Network Tokens, which also uses a decentralised, Ethereum based platform. Like most P2P models, providers and requestors are kept in check by their reputation. If providers send inaccurate computation results, or requestors fail to pay, then they lose reputation and are unlikely to be chosen again. To ensure security, computations take place in virtual machines which are isolated from the host’s system. The growth of the network demonstrates the expansion of sharing economy principles, as well as consumer support for decentralisation. The company’s aim is to become the first worldwide supercomputer, providing the necessary computing resources for any task. Their proof of concept, Brass Golem, was launched in August last year. There are three more planned versions yet to come – Clay, Stone, and Iron.

How will decentralised networks disrupt computing?
P2P services like Golem have the potential to make high computing power more accessible than ever, enhancing the ability of developers, programmers and designers to create quality content. Decentralised networks are also cheaper than traditional methods of accessing more power. Amateur developers will therefore be able to use them to their advantage as well. This may encourage a level playing field between seasoned and aspiring developers by making development more about skill than how much money is invested in high powered computer systems. As more people use P2P networks, the sharing economy will continue to expand. Whilst this will benefit requestors and providers (not to mention the team behind the service), it won’t be so great for existing options that are bound to lose a proportion of their customers. But, like the incumbent businesses surviving the rise of the gig economy, tried and trusted services will always draw consumers. The concept of shared computing power is still relatively new, and it will take time for people to adopt it. Nonetheless, P2P networks present a serious challenge to more expensive, centralised alternatives. The fact that payments are made via cryptocurrency is yet another indicator of the importance of digital currencies in finance, too.

The creation of a shared computer network bodes well for a future of accessible, quality development. The willingness of consumers to tap into the network demonstrates the growing popularity of decentralised systems, which is also shown by increasing interest in cryptocurrencies. Whilst existing, centralised systems have the advantage of longevity, Golem – and no doubt other similar networks – are cheaper and less binding. As the world is relentlessly digitalised, a greater number of programmers and developers will be needed to support and navigate digital disruption. With this in mind, perhaps the adoption of highly accessible computing networks is a smart move.

Will Golem become the first worldwide supercomputer? Are decentralised models the future of service provision? Should centralised services feel threatened by these new business models? Share your thoughts and opinions.