Interleaving read and write requests to reduce latency and maximize throughput in a flash storage device

摘要: In an all-flash storage array, write requests can take about 9 to 10 times longer than a read request of the same size. There could be several problems when reading or writing from storage, such as large slowing down small requests, other requests. Also, may slow smaller by filling incoming queue. one implementation, determination is made on what maximum size flash that improves throughput chip (e.g., beyond certain do not improve throughput). A chunklet defined block data having calculated As come in, are broken into chunklets, and then chunklets queued for processing chip. One processed at time per request. This way, does monopolize use period time, allowing while being storage.