Reading of distributed erasure-coded data from an enterprise object storage system
Abstract:
Various implementations disclosed herein enable reading of an erasure-coded file striped across one or more storage entities of an enterprise object storage system. For example, in various implementations, a method of reading an erasure-coded file is performed by an ingest entity of a storage system that includes a cluster of storage entities. The ingest entity includes a non-transitory computer readable storage medium, and one or more processors. In various implementations, the method includes querying the storage entities of the cluster to determine where data segments of an erasure-coded file are located within the cluster. In various implementations, the method includes generating a mapping of the data segments across the storage entities. The mapping indicates a sequence for the data segments within the erasure-coded file. In various implementations, the method includes scheduling read requests for the data segments based on the sequence. The scheduling satisfies a storage utilization threshold.
Information query
Patent Agency Ranking
0/0