Abstract:
A system and method for adaptive data placement in a distributed file system is provided. Upon creation of a new subdirectory, a first heuristic procedure is formed to determine whether the newly created subdirectory should be created locally or on a remote flexible volume. Should it be determination be made to create or store the subdirectory on a remote flexible volume, a second heuristic procedure determines which of a plurality of flexible volumes should store the newly created subdirectory.
Abstract:
A system and method caches and distributes meta-data for one or more data containers stored on a plurality of volumes configured as a striped volume set (SVS) and served by a plurality of nodes interconnected as a cluster. The SVS comprises one meta-data volume (MDV) configured to store a canonical copy of certain meta-data, including access control lists and directories, associated with all data containers stored on the SVS, and one or more data volumes (DV) configured to store, at least, data content of those containers. In addition, for each data container stored on the SVS, one volume is designated a container attribute volume (CAV) and, as such, is configured to store (“cache”) a canonical copy of certain, rapidly-changing attribute meta-data, including time stamps and container length, associated with that container.
Abstract:
Presented herein are methods, non-transitory computer readable media, and devices for efficiently repairing tree databases with variable-length records. Methods for repairing tree databases with variable-length records are disclosed which include: iterating the tree database, detecting at least one recoverable leaf block, reducing the at least one recoverable leaf block to its legally formatted contents, and writing the at least one recoverable leaf block back into the tree database; processing the at least one recoverable leaf block to ensure the at least one recoverable leaf block represents a disjoint region within a record space within the tree database; and rebuilding branch blocks as necessary to provide a lookup and organizational index for the at least one recoverable leaf block.
Abstract:
A system and method for adaptive data placement in a distributed file system is provided. Upon creation of a new subdirectory, a first heuristic procedure is formed to determine whether the newly created subdirectory should be created locally or on a remote flexible volume. Should it be determination be made to create or store the subdirectory on a remote flexible volume, a second heuristic procedure determines which of a plurality of flexible volumes should store the newly created subdirectory.
Abstract:
Methods, non-transitory machine readable media, and computing devices that more efficiently and effectively manage storage quota enforcement are disclosed. With this technology, a quota ticket comprising a tally generation number (TGN) and a local allowed usage amount (AUA) are obtained. The local AUA comprises a portion of a global AUA associated with a quota rule. The local AUA is increased following receipt of another portion of the global AUA in a response from a cluster peer, when another TGN in the response matches the TGN and the local AUA is insufficient to execute a received storage operation associated with the quota rule. The local AUA is decreased by an amount corresponding to, and following execution of, the storage operation, when the increased local AUA is sufficient to execute the storage operation.
Abstract:
Methods, non-transitory machine readable media, and computing devices that more efficiently and effectively manage storage quota enforcement are disclosed. With this technology, a quota ticket comprising a tally generation number (TGN) and a local allowed usage amount (AUA) are obtained. The local AUA comprises a portion of a global AUA associated with a quota rule. The local AUA is increased following receipt of another portion of the global AUA in a response from a cluster peer, when another TGN in the response matches the TGN and the local AUA is insufficient to execute a received storage operation associated with the quota rule. The local AUA is decreased by an amount corresponding to, and following execution of, the storage operation, when the increased local AUA is sufficient to execute the storage operation.
Abstract:
A system and method for implementing a quota system in a distributed file system is provided. Each node manages a quota database tracking available quota for the node. Should additional quota be required, a node queries a remote node to obtain a lock over the remote quota database. The additional quota is shifted and remaining free quota is reallocated between the local and remote nodes.
Abstract:
A system and method for adaptive data placement in a distributed file system is provided. Upon creation of a new subdirectory, a first heuristic procedure is formed to determine whether the newly created subdirectory should be created locally or on a remote flexible volume. Should it be determination be made to create or store the subdirectory on a remote flexible volume, a second heuristic procedure determines which of a plurality of flexible volumes should store the newly created subdirectory.
Abstract:
A system and method for data replication is described. A destination storage system receives a message from a source storage system as part of a replication process. The message includes an identity of a first file, information about where the first file is stored in the source storage system, a name of a first data being used by the first file and stored at a first location of the source storage system, and a fingerprint of the first data. The destination storage system determines that a mapping database is unavailable or inaccurate, and accesses a fingerprint database using the fingerprint of the first data received with the message to determine whether data stored in the destination storage system has a fingerprint identical to the fingerprint of the first data.
Abstract:
A system and method for implementing a quota system in a distributed file system is provided. Each node manages a quota database tracking available quota for the node. Should additional quota be required, a node queries a remote node to obtain a lock over the remote quota database. The additional quota is shifted and remaining free quota is reallocated between the local and remote nodes.