Читать книгу Blockchain Data Analytics For Dummies - Michael G. Solomon - Страница 45

The last question in the preceding section leads well into one of the biggest current obstacles to blockchain adoption. Scaling performance to an enterprise scale is an ongoing pursuit that hasn’t been completely resolved. Most enterprise applications use legacy database management systems to store and retrieve data. These data repositories have been around for decades and have become efficient at handling vast amounts of data.

According to Chengpeng Zhao (CEO of the cryptocurrency exchange Binance), a blockchain implementation must be able to support 40,000 transactions per second to be viable as a core technology in a global cryptocurrency exchange. Currently, only four popular blockchain implementations claim to be capable of more than 1,000 transactions per second (Futurepia, EOS, Ripple, and NEO). The most popular public blockchain, Ethereum, currently can handle about 25 transactions per second. Future releases of Ethereum, however, are focusing on raising the transaction throughput substantially. The technology is getting better but has a long way to go to be ready for the volume that enterprises require.

Performance isn’t the only limiting factor when assessing blockchain for the enterprise. Integration with legacy artifacts and the ease with which the blockchain infrastructure fits into the existing enterprise IT infrastructure are concerns as well. Do all blockchain nodes require new virtual or physical hardware? Can the new nodes run on existing servers? What about network connectivity? Will existing network infrastructure support the new blockchain network? These are only a few of the many questions that enterprises must answer before deploying a blockchain integration project.