摘要
早期区块链系统主要运行在冯·诺依曼架构的通用处理器上,随着区块链技术的发展,其在各行各业得到广泛应用和大规模部署的情况下,在计算密集型和通信密集型场景中,需要同时兼顾高计算能效和高计算灵活性。本文在此基础上,利用本地计算机和云服务器这类通用处理器结合现场可编程门阵列(FPGA)组成异构计算区块链网络,从而经济有效地获得了高能效计算能力、优秀的兼容性和可拓展性,以及较高的计算资源利用率,为区块链计算方式由同构走向异构提供底层技术方案,带来了更多的区块链应用可能性,具有广阔应用前景。
Early blockchain systems mainly run on general-purpose processors with von Neumann architecture such as Central Processing Units(CPUs)and Graphics Processing Units(GPUs).With the development of blockchain technology,it has been widely used and deployed on a large scale in various industries.In computing-intensive and communication-intensive scenarios,both computing energy efficiency and high computing flexibility need to be considered.On this basis,this article uses general-purpose processors such as local computers and cloud servers to form a heterogeneous computing blockchain network,thereby cost-effectively obtaining high-energy-efficiency computing capabilities,excellent compatibility and scalability,etc.The high utilization rate of computing resources provides an underlying technical solution for blockchain computing fr om homogeneous to heterogeneous,which brings more possibilities for blockchain applications and has broad application prospects.
作者
徐易朗
毕涛
赵建业
XU Yilang;BI Tao;ZHAO Jianye(School of Electronics Engineering and Computer Science,Peking University,Beijing 100871,China)
出处
《太赫兹科学与电子信息学报》
2021年第4期697-704,共8页
Journal of Terahertz Science and Electronic Information Technology
基金
国家自然科学基金资助项目(91836301)
关键词
区块链
异构计算
现场可编程门阵列
计算能效
blockchain
heterogeneous computing
Field-Programmable Gate Array
computing energy efficiency
