High Available Data Storage Systems
Data Storage is the bottleneck of today information systems stemming from high latency of current storage media. Rapid growth of market needs for high-dependable, high-throughput, and fast storage approaches the information infrastructures to employ Data Storage Systems (DSS), simultaneously serving tens of requests through ultra fast communication ports. DSS dependability is dominantly threatened by Data Unavailability (DU), Data Loss (DL), and Silent Data Corruption (SDC) events. DSS dependability threats can have many causes such as Single Event Upset (SEU)s, hardware failures, double (multiple) disk failures, memory module failures, Latent Sector Error (LSE)s, and power outages. Data availability is one of the most important properties of interest that a DSS must provide. For data management businesses, each hour of downtime costs tens of thousands of dollars. Consequently, reducing the downtime per year is a principle for obtaining an optimal Total Cost of Ownership (TCO) in IT businesses. To this end, system design should be aware of sources of failures and dependability bottlenecks to reach the desired dependability at the minimum redundancy and cost.
Low-Power Error Detection and Correction Codes in Microprocessor Cache Memory
Using Error Detection Code (EDC) and Error Correction Code (ECC) is a noteworthy way to increase cache memories robustness against soft errors. EDC enables detecting errors in cache memory while ECC is used to correct erroneous cache blocks. ECCs are often costly as they impose considerable area and energy overhead on cache memory. Hence, reducing this overhead is a hot topic in the computer archtecture field and have been the subject of many studies.