CMSS: A High-Performance Blockchain Storage System with Horizontal Scaling Support

Author:

Yang Wenjin12,Ao Meng2,Gao Mingzhi1,Li Chunhai3,Chen Yongqing2

Affiliation:

1. School of Cyberspace Science & Technology, Beijing Institute of Technology, Beijing 100081, China

2. Tencent Inc., Shenzhen 518055, China

3. Guangxi Engineering Research Center of Industrial Internet Security and Blockchain, Guilin University of Electronic Technology, Guilin 541004, China

Abstract

As a decentralized system, blockchain has been widely used in numerous areas and has become a hot topic in both industry and academia. The increasing demand for blockchain causes heavy storage consumption which seriously affects the performance of blockchain, especially in the context of massive volumes of data. To solve these problems, many related systems like sharding and sidechain have been proposed to improve the efficiency and scalability of blockchain. However, in practical deployment scenarios, these systems still have problems, such as low read–write performance, and reorganization and synchronization of ledger data after storage expansion, which cause the storage system’s expansion to become difficult and time-consuming in large-scale blockchain systems. Facing these problems, in this paper we propose ChainMaker Storage System (CMSS). CMSS is a blockchain storage system with high read-and-write performance and horizontal scaling support. It has been used as the storage system of the most popular permissioned blockchain ChainMaker. There are three contributions of our proposed CMSS: (i) a new block storage workflow to achieve high read–write performance; (ii) the Meta File System (MFS) to support the horizontal expansion of blockchain storage; and (iii) hot–cold separation to reduce the resource usage and economic costs. To evaluate the performance of CMSS, we compare CMSS with Hyperledger Fabric (HLF), the most popular permissioned blockchain platform. We select five well-known cloud service providers to calculate the storage cost in a real production environment. The results show that CMSS has better performance in read and write than HLF and advantages in storage capacity and price.

Funder

National Key Research and Development Program of China

National Natural Science Foundation of China

Publisher

MDPI AG

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3