After the recent cryptocurrencies’ crash, we are seeing a higher concentration of mining power. A problem already discussed in many articles, which found a huge Chinese predominance giving the cost of technology and electricity in the second global economy.
We won’t spend any words about the 51% attack problem, we will mainly focus on energy supply problem, as China suffered many energy outages during its extremely rapid growth, and their effect was extensively studied in economic literature. In a recent study, Cheng, Wong and Woo have been very cautious about the ability of China to face with its difficulties: “Given China's past and current shortages, however, an increase of 1% in electricity generation will not, in the next few years, achieve the generation reliability target typically used in OECD countries [A commonly used reliability target is of one day in 10 years, implying a reserve margin of 15–25% of system peak demand. In contrast, areas around Shenzhen in the Guangdong province of South China had, in the summer of 2011, planned outages of one day per week for industrial users]”. As a consequence, in august 2017, Shenzhen witnessed many outages caused by the heat wave.
Bloomberg recently highlighted some threats and described the chinese energy hunger: “In the first half of the year, power demand from agriculture, manufacturing and service industries climbed by 10 percent, 7.6 percent and nearly 15 percent, respectively, according to the China Electricity Council. Residential use increased by 13 percent, it was said last week. The country’s total power consumption increased by 9.4 percent between January and June, NDRC’s Yan said. That’s the highest percentage since 2011, according to data from the National Energy Administration”.
As blackout is not so rare, we need to conceive a way in order to secure electric supply to mining farm, that, incidentally, are very energy demanding. In fact, a long lack of mining power could lengthen verification time, making transaction much longer and even uncertain, in the worst case. Even if it is possible to adapt hash rate to the available hash power, this is not a fast solution, because it is not compatible with the extreme rapidity, and imprevedibility of a black-out.
Although the blockchain node architecture for bitcoins involves the use of full and light nodes with a remarkable difference in storage and management of information (full nodes make a copy of the whole chain or blocks and transactions and make an important contribution to check all the rules of the blockchain while the most numerous light nodes have a simplified check operation and do not need to memorize blocks and transactions), it is good to share some numbers: China's full-nodes are less than eight percent of the almost ten thousand present worldwide but the overall hash rate provided by global Chinese mining exceeds the seventy percent.
What is the solution?
In order to rightly conceive the threats, we divide the problem in two: the grid reliability and the source of energy. As we have already seen, grid is exposed to excessive demand, technical problems connected to the weather (as, for example, heat impacts on wires and fuses) and policy’s change in electric priority access; usually, industry is the first to stop its huge consumption when the system is going close to the peak.
Moving now to analyze the source’s related aspects, the cheapest one is coal, which is a very polluent form of energy, and may one day be subject to some kind of “green taxation”. Luckily, a lot of miners employ direct sources, like hydropower. A recent news about the crackdown decided by Sichuan Electric Power Company cast some shadows also about this supply.
Furthermore, we need to consider all the remote risks: hydropower suffers from climate change, as a lack of rain could lead to close the implants. In addition, during winter, many rivers are frozen.
What real impact could one have on the timing of transaction verification? It is important to underline that we are not talking about the impact on the solidity of the architecture but of its performances commensurate with the current ones.
Some indirect solutions that look at energy sustainability and a democratic distribution of the mines have been found (see the proof of stake), but it is also necessary to regulate miners with energy best practices to guarantee a "long term proof of work". The problem is not to be entirely underestimated and it has also an impact on energy policy, as currently the energy consumption of miners tends to increase as the algorithm that underpins the PoW leads miners to invest more and more money in power calculation.
Our proposal is to create some sort of self-regulation to which the miners should comply; here is a sketch of the aspects to watch out for:
About energy and business continuity:
- use of a first reliable source;
- availability of a second source of energy, completely independent, namely a UPS (uninterruptible power supply);
- availability of hash power from different areas (it is a good news that Bitmain is investing in Texas).
About compliance, the simplest requirements are:
- offer clear answers to business partners who wish to know more about energy power reliability;
- publish on their website all the relevant info on the plans of supply and business continuity [granting, of course, the need to hide some info for commercial reasons];
- publish annually info on the accidents suffered.
Choices on UPS (battery / inverter) or fuel generators are not simple and must be calibrated according to the required duration, power, budget and logistics availability. In general, the investment for the purchase of a UPS for small miners is not to be underestimated and can be expensive and with a long break-even over time. In these cases, the purchase of a fuel generator could be a good choice. Large plants, on the other hand, require investments towards UPS and in structured and industrialized cases, some facilities have invested in large plants close to energy plants).
There are also on the market UPS that allow the battery to be charged constantly from renewable energy sources thus allowing both redundancy and significant energy savings; all these are very important aspects, since the possibility for miners to gain is closely linked to the trade-off between the reward of the cryptocurrencies obtained with respect to energy consumption.
We think that our idea has a good recent example of the chance of self-regulation, as in the US, where four exchanges have created an association with the aim of submitting to their own regulation, waiting the moves of the public regulators.