Efficient High-Performance FPGA-Redis Hybrid NoSQL Caching System for Blockchain Scalability
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review
Author(s)
Related Research Unit(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 81-91 |
Number of pages | 11 |
Journal / Publication | Computer Communications |
Volume | 169 |
Online published | 27 Jan 2021 |
Publication status | Published - 1 Mar 2021 |
Link(s)
Abstract
The inspiring blockchain technology has got successes and many adoptions in areas besides cryptocurrencies as its benefits have been explored and successfully tested. Scalability is one of the biggest challenges of blockchain. Due to the huge increasing size of blockchain data, many devices (lightweight nodes) especially IoT depend on full blockchain servers, hence there is a need to reduce the workload on the servers for high performance. This paper presents a high performance and efficient hybrid (multilevel) and distributed NoSQL caching system of FPGA and Redis for improving the scalability (throughput) of blockchain applications. We investigate performance bottlenecks in blockchain and design an efficient Gigabit Ethernet FPGA NoSQL cache architecture that works in synergy with Redis database via Hiredis C client. Curl and Jansson are used to connect with the blockchain. We design a customized SHA-256 core for the efficient caching specific for blockchain. Our results revealed an improvement of 103 times when cache hit occurs on the FPGA without cache miss. When cache miss occurs on the FPGA, up to 4.09 times improvement is obtained with the proposed FPGA–Redis system compared to FPGA only system. Small FPGA area utilization and less power consumption were also achieved.
Research Area(s)
- Blockchain, Caching and Throughput, FPGA, Redis, Scalability
Citation Format(s)
Efficient High-Performance FPGA-Redis Hybrid NoSQL Caching System for Blockchain Scalability. / Sanka, Abdurrashid Ibrahim; Chowdhury, Mehdi Hasan; Cheung, Ray C.C.
In: Computer Communications, Vol. 169, 01.03.2021, p. 81-91.
In: Computer Communications, Vol. 169, 01.03.2021, p. 81-91.
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review