SBK: A Framework for Performance Benchmarking for a Variety of Storage Systems

Abstract

Benchmarking storage systems at scale can be challenging. Within the realm of big data, performance stands out as a significant challenge. Proper storage and maintenance of big data are crucial to guarantee accessibility, achieve cost savings, enhance risk management, and gain a deeper comprehension of customer needs. This paper addresses the challenges faced in managing extensive and rapidly growing data volumes and emphasizes the importance of maintaining optimal storage performance. The SBK framework is containerized and vendor-neutral, making it easy to use and deploy. It is a software benchmarking framework designed to evaluate the performance of any storage system, inclusive of all types of data/payload. This paper demonstrates the use of SBK in benchmarking and highlights the relevance of benchmark testing in evaluating storage performance. SBK aims to provide transparency and ease of use for benchmarking purposes. This framework functions correctly with different hardware configurations, operating systems, and software environments.