Storage Reliability

The rapid development of information technology has put data at the heart of modern production, and data is now experiencing a period of tremendous growth. In 2019, the International Data Corporation (IDC) predicted that the Global Datasphere (a measure of all new data that is captured, created, and replicated in any given year across the globe) will grow from 32 ZB (zettabytes) recorded in 2018 to 175 ZB by 2025 [1]. Traditional data centers match the storage demands stemming from all this data creation either by beefing up installed devices or setting up extra devices. But a storage system with more devices succumbs more easily to failures and disruptions. Recently, over half of data centers have reported service or access disruptions, each incurring million-dollar losses or more. In light of these risks, a data storage system must provide high reliability through robust fault tolerance policies and mechanisms that are designed based on the hardware and software failure modes.