Methods for storing data in a data storage system where a RAID-X format or formats are implemented at a file level
View Patent ↗A data storage system where a RAID-X format is implemented at a file level includes storage devices, and one or more nodes that access files from the storage devices. N represents a maximum limit of failures beyond which reconstruction of a file stored using the file level implementation is not possible. A directory is encoded for each of a plurality of files, and a map is encoded that includes a list of storage devices where components of the directory associated with the map reside. The encoding of the directory is performed using a RAID-X format that can tolerate at least N+1 failures prior to being unable to reconstruct the directory.
1. In a data storage system that includes a plurality of storage devices, and one or more nodes that access files from the storage devices, wherein a RAID-X format is implemented at a file level within the plurality of storage devices, and wherein N represents a maximum limit of failures beyond which reconstruction of a file stored using the file level implementation of the RAID-X format is not possible, a method comprising:
encoding a directory for each of a plurality of files stored on the storage system, and a map that includes a list of storage devices where components of the directory associated with the map reside; and
wherein the encoding of the directory is performed using a RAID-X format that can tolerate at least N+1 failures prior to being unable to reconstruct the directory; and
wherein the data storage system is a distributed, object-based data storage system, and each of the storage devices is an object-based storage device.
2. In a data storage system that includes a plurality of storage devices, and one or more nodes that access files from the storage devices, wherein a RAID-X format is implemented at a file level within the plurality of storage devices, a method comprising:
storing a directory (D) on a set of storage devices such that no file or subdirectory contained in D has associated metadata on the exact same set of storage devices where D is stored; and
wherein metadata of each child of D includes a back-pointer to D that identifies at least a parent directory and a filename in said parent directory;
whereby, if all storage devices in the set used to store D fail, all children of D have at least one surviving metadata component which can be used to identify locations in a filesystem namespace associated with the data storage system of all of said children.
3. The method of claim 2 , wherein the data storage system is a distributed, object-based data storage system, and each of the storage devices is an object-based storage device.