The public was aghast back in early 2018 when PwC Bitcoin specialist Alex de Vries estimated that Bitcoin servers consume more than 22 terawatt-hours of electricity per year which is comparable to entire countries such as Austria and Ireland.
To many, this was a brewing crisis that if left unchecked would render blockchains unscalable and unsustainable Even Bitcoin’s strongest supporters recognized that the energy consumption issue presented a barrier to growth and mainstream adoption. The activity that consumes so much energy is called mining. Mining is the lifeblood of all Proof-of-Work (POW) based cryptocurrencies, of which Bitcoin is the largest.
When miners “mine” they are supporting the blockchain network by validating transactions and forming new blocks to be add to the blockchain. They do this by solving complex cryptographic puzzles. The first miner to solve this puzzle validates a new block and earns a block reward. Earning this block reward has become increasingly competitive over time. POW algorithms have a difficulty mechanism that throttles how quickly new blocks can be mined. This throttling mechanism makes the cryptographic puzzle more difficult as more computational power is added to the network to keep the time between newly created blocks approximately constant. The only way existing miners can stay competitive with the additional computational power is by bringing on more computational power of their own. This creates an arms race and an industry with runaway energy consumption.
But at some point there has to be a limit. There has to be a limit to how much hardware and electrical power can be thrown at a POW network and it still making economic sense. Well, there are some indications that we may be approaching that limit. Nvidia, a Taiwan-based producer of graphics processing units (GPU) was the worst performing company in the S&P 500 in Q4 2018 after its stock fell 54 percent. Jensen Huang, the company’s CEO, cited what he called a “crypto hangover” as blockchain related hardware sales declined.
There are two headwinds facing blockchain related hardware producers. The first is the current bear market in cryptocurrency prices which affected the stock prices of hardware makers like Nvidia, AMD and others. The second is the more long term trend toward Proof-of-Stake (POS) protocols.
POS protocols will make the mining process ‘virtual’, effectively replacing miners with validators. The main problem POS protocols solve is they will greatly reduce the energy consumption required to earn block rewards. Instead of the block reward being competitive, a function of the amount of capital a miner invests in hardware and electrical energy; in POS protocols, the block reward is a function of how much the validator stakes or the amount of coin the validator offers as collateral. Therefore, in order to earn more rewards, the validator buys and stakes more coins, instead of adding more hardware and consuming more electricity.
Ethereum, the world’s second largest cryptocurrency, has announced that their POS implementation will be called the Casper Protocol. While Ethereum has not announce an official launch date for Casper yet, Ethereum has made it clear that there is a real “need” for POS protocols.
As traditional POW mining becomes less and less profitable and as proof of stake consensus mechanisms become more prevalent, the next logical question is ‘What will happen to all of the mining rigs?’
In 2014, iDrive conducted an experiment to see if excess computational capacity in corporate data centers could be used to mine Bitcoin. It worked! 600 servers run 24/7 for a year yielded 0.4 BTC. While that isn’t very impressive compared to ASICs (Application Specific Integrated Circuits) that are specifically built for mining, it did open the door for a more interesting experiment.
What would happen if you pointed hardware currently designated for mining at a cloud compute solution? The industry is still in its very early stages but Blockchain based compute and storage solutions like Golem and Sia seem to be very promising. Right now, the market rate for storage on Sia is orders of magnitude less expensive than traditional cloud storage solutions like Amazon Web Services.
The reason why blockchain solutions like Sia can be so much cheaper than AWS is partly because of the difference in hardware but mostly because of an economic phenomenon known as ‘Uberization’ or more formally known as collaborative consumption.
Collaborative consumption, or the sharing economy, refers to an ecosystem where consumers can potentially have a two-sided role, serving as both obtainers of resources and providers of those same resources. On-demand transportation companies like Lyft and Uber (hence the name, Uber-ization) allow individuals to hail their ‘peers’ for a ride or they can also lend their personal time and vehicle back to the community and give others a ride. Similarly, AirBNB and WeWork allow property owners the opportunity to ‘share’ available real estate to those who need it.
The economic forces that drive certain industries toward a collaborative consumption model is the opportunity to better optimize resources by matching demand to underutilized or excess capacity. For example, the third-party logistics company, CH Robinson has built a $15B enterprise by matching freight demand to truck drivers with excess capacity. After delivering a load in one part of the country, truck drivers are often willing to pickup and deliver a load close to their home at a discount in exchange for not having to drive all the way home with an empty trailer.
The industry best positioned right now for disruption through collaborative consumption is cloud computing. While computers and devices have become more powerful largely following Moore’s Law and the cost of electronics has generally been falling, the cost of cloud computing has curiously remained relatively steady and in some cases has increased. This is because the cloud computing market is not facing the normal pricing pressures one would expect from a competitive market. Cloud computing is currently dominated by a small number of huge multinational software conglomerates that are consolidating the industry and moving closer and closer to monopolistic pricing.
The spread (or difference) between how much these software conglomerates are charging for cloud computing services and the cost blockchain miners can provide the same services is pretty large and will benefit consumers greatly. The potential profits miners will be able to earn will be much larger than when competing against each other for diminishing block rewards. The cloud computing industry will benefit greatly from blockchain’s innate democratization of resources but what can’t be overlooked will be the dramatic increase in stability and security.
But every economic innovation depends on technological enablers to initiate the shift. Uber was able to revolutionize the ride hailing market just when Smartphones reached critical adoption, and enabled millions of customers to connect to their services. And while digital wallets are now readily available for many blockchains there is still a lack of availability to access uberized cloud infrastructure through blockchain technology. The necessary integrations to connect existing appliances, IT infrastructure or IoT devices and solve business problems better or cheaper is just not there.
Exosite, an industrial IoT company is contributing to make this change happen. Leveraging blockchain technology we have built a technology stack specifically designed to secure machine to machine communications. Diode, is a public blockchain and an IoT device stack that puts an end to man-in-the-middle attacks and acts as a registry for digital certificates and authorities. It enables individuals and corporations to leverage blockchain services to send, broadcast, store, and proof data in smooth integration with existing infrastructure. If you’re managing a device fleet looking to offload expensive infrastructure costs or running a mining operation feel free to reach out to us and learn more about our partner program.