Blockchain technology emerged from not-so-new concepts such as cryptography and distributed systems. However, are blockchain-based applications suitable to replace traditional distributed applications? Performance evaluation can help us to achieve an answer to this question. This paper assesses the current state-of-the-art performance evaluation in some of the most popular platforms that provide public, private, or cross-operation blockchain-based applications. We highlight key studies, technologies, and metrics focusing on three main research aspects: (i) Scalability, (ii) Security, and (iii) Interoperability. Some of the key obtained results point out that works regarding interoperability are an important gap in the blockchain performance field. Hyperledger Fabric and Ethereum are among the most prominent platforms, meaning that academics and industry need to explore other platforms. Finally, we discovered that measurement was the most used performance evaluation method, and it mainly focused on transaction latency and throughput.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Blockchain Review Based on the Scalability Trilemma

  • Carlos Melo,
  • Glauber Gonçalves,
  • Francisco A. Silva,
  • André Soares

摘要

Blockchain technology emerged from not-so-new concepts such as cryptography and distributed systems. However, are blockchain-based applications suitable to replace traditional distributed applications? Performance evaluation can help us to achieve an answer to this question. This paper assesses the current state-of-the-art performance evaluation in some of the most popular platforms that provide public, private, or cross-operation blockchain-based applications. We highlight key studies, technologies, and metrics focusing on three main research aspects: (i) Scalability, (ii) Security, and (iii) Interoperability. Some of the key obtained results point out that works regarding interoperability are an important gap in the blockchain performance field. Hyperledger Fabric and Ethereum are among the most prominent platforms, meaning that academics and industry need to explore other platforms. Finally, we discovered that measurement was the most used performance evaluation method, and it mainly focused on transaction latency and throughput.