Environmental and Human Costs of Bitcoin and the Blockchain

From P2P Foundation
Jump to navigation Jump to search


The tremendous Environmental and Human Costs of Bitcoin and the Blockchain

Darin Stevenson:

"What is bitcoin, really? You can think of it as a machine contagion—a network of devices amped to their performance tolerances—machines that do nothing but reprocess every transaction that ever occurred on their network (thus achieving ‘consensus’) while, at the same time, ‘mining’ new blocks of coins (currently worth about 25 btc) by solving a purposefully cumbersome mathematical formula which doesn’t actually accomplish anything other than enforcing computational difficulty. That is: making millions of machines grind away madly at nothing.

This mining process is both the reprocessing of transactions, and a ‘weight’ factor that is incremented to insure that the average time for the entire network to ‘solve’ a block (that is, to produce an accurate guess close enough to a mathematically supplied target) is ‘about 10 minutes’. When you initiate a new machine into the network, you download the current transaction record (a 6-gigabyte file) of the entire history of bitcoin and reprocess it (this takes around 24 hours). You then either ‘mine’ alone (an almost useless endeavor which would take ~98 years to solve a block) or you join a pool of machines. By joining a pool, you get statistically better performance in terms of satoshi (currently: USD $0.0000046543 each) earned as you are ‘rewarded’ for work done by your machine’s participation in the pool.

The more processing power (raw computational force over time) you can bring to bear on ‘the problem’ … the better a chance you have to earn incremental additions to your ‘wallet’ or account. Of course, most of the problem is invented… to be this kind of problem—one that requires more and more computational activity to qualify as complete. And we have now invented specialized machines and chips just to solve this problem.

Machines involved in mining are pressed to the limits of their power consumption and performance profiles; they are ‘pinned’ at 100%+ of their computing power, ceaselessly, and thus generate heat (as well as consuming copious quantities of electrical power).

Because they remain hot, they have to be electronically ventilated. This process of power-heating something we must in turn power-cool, for the phony ‘sake’ of mathematical processes intended to make more work each time they are implemented is deranged. We’re essentially turning computers into heaters that we have to cool to recompute previous computations with. On purpose. A single day of the environmental costs of this process are so catastrophic that if we ever did the accounting — if anyone did — we would immediately understand that this entire idea is a mode of ‘fracking the whole environment’ whose costs rise explosively with every moment we continue the process. Only a species that had gone entirely insane, and consciously intended to wipe out life on Earth would ever consider such a process. But any species that could actively celebrate and expand it — must be understood as both emphatically suicidal and openly omnicidal. In short: they intend to kill everything, anything, and themselves — and are hell-bent on the continuous and unlimited expansion of this agenda.

The activity of the blockchain networks are not only burning down the future; they are also obliterating the history of humanity and life on Earth, faster and more aggressively every moment, by destroying the living results and opportunities established by this history, and insuring that the benefits that might otherwise blossom into astonishing ‘interest on investment’, are killed off by the necessity of breeding, operating, heating and cooling millions of machines that do makework for a resource that only exists as numbers in machines." (https://medium.com/ill-ixi-lli/-da9326e97d1f)

Energy Usage Aspects

Bitcoin's Energy Use

1. Brad Plumer:

"Blockchain.info, a site that tracks data on Bitcoin mining, estimates that in just the last 24 hours, miners used about $147,000 of electricity just to run their hardware, assuming an average price of 15 cents per kilowatt hour … That’s enough to power roughly 31,000 U.S. homes, or about half a Large Hadron Collider.

It’s a stunning stat, but does this really count as a “disaster”? That’s less clear. After all, we need to consider the counterfactural: Is it possible that these computers would be used for other activities and calculations anyway, if they weren’t mining Bitcoins?

In any case, Gimein’s piece does touch on a red-hot topic in energy circles — how much electricity does all of our computing and Internet infrastructure actually consume? A 2011 study by Stephen Ruth of George Mason University estimated that the entire global information and communications technology industry accounts for “only about 3–5 percent” of the world’s electricity use. So it has a much smaller environmental footprint than, say, cars, trucks, and planes (which account for 25 percent of all energy demand.)

On the other hand, the Internet’s energy needs are expected to swell significantly in the coming years — even though computing keeps getting more energy-efficient. An interesting new study in Science by Diego Reforgiato Recupero finds that Internet traffic volume tends to double every three years. But network energy-efficiency isn’t keeping pace. As a result, the world’s IT infrastructure will consume 19 percent more energy in 2013 than in 2012.

Interestingly, as Alexis Madrigal explains here, most of the energy used by our computing infrastructure comes from wireless and cellular networks — by contrast, data centers themselves only use about 10 percent of the electricity involved. What’s more, those wireless networks don’t seem to be improving their energy efficiency all that quickly. That’s why overall energy use could keep growing, particularly as cloud computing becomes more widespread.

Bottom line: On the vast scale of environmental disasters, Bitcoin barely registers. And, in the grand scheme of things, the Internet is still relatively green (that’s particularly true if it cuts into other activities, like driving). But it’s also true that our computing infrastructure is becoming an increasingly significant part of the world’s energy demand." (http://www.washingtonpost.com/blogs/wonkblog/wp/2013/04/12/what-bitcoin-teaches-us-about-the-internets-energy-use/)

2. Eric Limer:

"According to Bitcoin Watch, the whole Bitcoin network hit a record-breaking high of 1 exaFLOPS this weekend. When you're talking about FLOPS, you're really talking about the number of Floating-point Operations a computer can do Per Second, or more simply, how fast it can tear through math problems. It's a pretty common standard for measuring computer power. An exaFLOPS is 1018, or 1,000,000,000,000,000,000 math problems per second. The most powerful supercomputer in the world, Sequoia, can manage a mere 16 petaFLOPS, or just 1.6 percent of the power geeks around the world have brought to bear on mining Bitcoin. The world's top 10 supercomputers can muster 5 percent of that total, and even the top 500 can only muster a mere 12.8 percent.

And that 1 exaFLOPS number is probably a little low. Because Bitcoin miners actually do a simpler kind of math (integer operations), you have to do a little (messy) conversion to get to FLOPS. And because the new ASIC miners—machines that are built from scratch to do nothing but mine Bitcoins—can't even do other kinds of operations, they're left out of the total entirely. So what we've got here is a representation of the total power spent on Bitcoin mining that could theoretically be spent on something else, like real problems that exist naturally.

Because of the way Bitcoin self-regulates, the math problems Bitcoin mining rigs have to do to get more 'coin get harder and harder as time goes on. Not to any particular end, but just to make sure the world doesn't get flooded with Bitcoins. So all these computers aren't really accomplishing anything other than solving super difficult and necessarily arbitrary puzzles for cyber money. It's kind of like rounding up the world's greatest minds and making them do Sudokus for nickels.

Projects like [email protected] and [email protected] use similarly networked power for the less-pointless practices of parsing information that could lead to more effective medicines or finding extra-terrestrial life, respectively, and either are hard-pressed to scrounge up even half of a percent of the power the Bitcoin network is rocking. And with specialized Bitcoin-mining hardware on the rise, there's going to be an army of totally powerhouse PCs out there that are good for literally nothing but digging up cybercoins.

It's incredible to think about the amount of power being directed at this one, singular purpose; power that's essentially being "donated" by thousands of people across the globe just because they have skin in the game. It's by far the most computational effort that has ever been devoted to a single purpose. And sure, Bitcoins are fine and all, but can you imagine what we could do if this energy was put behind other tough problems? We'll you're going to have to imagine, because so long as mining Bitcoins can earn you money and folding proteins can't, it's pretty clear which one is gonna get done." (http://gizmodo.com/the-worlds-most-powerful-computer-network-is-being-was-504503726?)


"At today’s value of roughly $1,000 per bitcoin, the electricity consumed by the bitcoin mining ecosystem has an estimated carbon footprint – or total greenhouse gas emissions – of 8.25 megatonnes (8,250,000 tonnes) of CO2 per year, according to research by Bitcarbon.org. That’s 0.03 percent of the world’s total greenhouse gas output, or equivalent to that of the nation of Cyprus. If bitcoin’s value reaches $100,000, that impact will reach 3 percent of the world’s total, or that of Germany. At $1 million – which seems farcical but which may not be out of the realm of possibility given the artificially limited bitcoin supply – this impact rises to 8.25 gigatonnes, or 30 percent of today’s global output, and equivalent to that of China and Japan combined.

Bitcoins aren’t mined from the earth’s crust like most physical commodities – although at least that leaves tangible evidence of its environmental impact. Rather, they are “mined” by computers solving a set of complicated computational problems. These problems are designed to get more difficult over time, until the year 2140 when the 21 millionth (and final) bitcoin is mined. Early in bitcoin’s existence, it was feasible to run a successful mining operation with a standard PC. Now the task requires custom mining rigs that can run orders of magnitude more processes per second.

The top of the line model, which is currently made by a Swedish company called KnCMiner, costs around $13,000 and can mine at a rate 550 gigahashes per second: They’ve sold $28 million worth, and soon these too will be obsolete. The total computational power of the global bitcoin mining network today is more than seven million gigahashes, and climbing. That’s 256 times greater than the world’s top 500 supercomputers, combined.

These computers are consuming so much electricity that it’s already unprofitable to mine in some regions of the world. According to Blockchain.info the total electricity cost of all mining acticity conducted over the last 24 hours was $19,652,986.38, as the system consumed 131,019.91 megawatt hours. In April, Bloomberg Sustainability called bitcoin mining it a “real-world environmental disaster.” At the time, the system was consuming just 7,000 megawatt hours per day – things have increased 142-fold in the last eight months." (http://pando.com/2013/12/16/bitcoin-has-a-dark-side-its-carbon-footprint/)