According to a large group of university experts, without a major redesign of the whole system, bitcoin would be unable to handle more than 27 transactions per second and remain decentralised.
Referring to the heated block size arguments across the bitcoin community, Ari Juels, a cryptography professor at Cornell Tech, claims that “current debate is missing the forest for the trees.” Together with a large team of researchers from six American universities, he authored a position paper On Scaling Decentralized Blockchains that will be presented at the 3rd Workshop on Bitcoin and Blockchain Research taking place in Barbados between 22 and 26 February.
Bitcoin in its current state is only able to process 7 transactions per second while such payment processor as Visa handles more than 2,000 transactions per second on the average, the peak capacity being up to 56,000. The problem is widely recognised, and many bitcoin improvement proposals have been submitted, including Bitcoin XT, Bitcoin Classic, Bitcoin Unlimited and Segregated Witness. Most of them suggest raising block size limit above the current 1MB.
Meanwhile, according to the experts, if bitcoin’s decentralised character is to be kept, the block size cannot be increased to more than 4MB and thus the maximum throughput of bitcoin will reach 27 transactions per second.
Researchers emphasise that they consider only some bitcoin metrics because not all can be accurately measured. That means that their preliminary conclusions are rather optimistic: “additional metrics could reveal even stricter limits.” If the results of the research are to be taken seriously, it would indicate that bitcoin as it is now, despite what its enthusiasts say, is hardly suitable for micropayments promoted by ChangeTip or 21 Inc., especially as the cost of one confirmed transaction, according to the paper, equals $1.4 - $2.9, where 57% goes for electricity consumption for mining.
The scholars suggest different directions to look for the solution “that will allow blockchains to scale beyond the parameters of today’s Bitcoin.” The stated goal is to “paint the design space, suggest promising approaches, and pose open challenges to the community.” They range from small adjustments to a radical redesign of the whole blockchain.
In August 2015, three of the authors of the current paper, Ari Juels, Elaine Shi and Ahmed Kosba, pubished a research article The Ring of Gyges: Using Smart Contracts for Crime.
Some of the team are members of the research project involving three American universities (Cornell University, the University of Maryland and the University of California, Berkeley) that were awarded a $3 million grant to explore cryptocurrencies and smart contracts.