International Business Machines Corporation

Vast Data Ltd.

Oded Sonin

Yariv Bachar

Ron Edelstein

Shay Haim Akirav

Asaf Levy

Lior Aronovich

Ron Asher

Avi Goren

Lior Klipper

Konstantin Mushkin

Ehud Meiri

Alon Horowitz

Ilai Harsgor-Hendin

Aviv Caro

Yogev Vaknin

Ido Yellin

Guy Keren

Hadar Hen-Zion

Ori Mamluk

Ido Yelin

Dror Dayan

A method for updating file system entity (FSE) metadata. The method may include receiving, by a first computing entity, a request to perform a case-sensitive FSE operation that comprises one of (a) adding a case-sensitive version of a name of a FSE to a data structure, or (b) deleting the case-sensitive version of the name of the FSE from the data structure, the data structure comprises multiple name blocks; finding, by the first computing entity, a certain name block of the multiple name blocks that is associated with the case-sensitive version of the name of the FSE. The method may also include determining, by the first computing entity, whether (a) the certain name block is associated with a case-insensitive version of the name of the FSE, and whether (b) the first computing entity preceded a second computing entity in reaching the certain name block, the second computing entity was requested to perform a case-insensitive FSE operation that corresponds to the case-sensitive FSE operation; when it is found that (a) the certain name block is associated with the case-insensitive version of the name of the FSE, and (b) the first computing entity preceded the second computing entity in reaching the certain name block, then: a) informing the second computing entity that the first computing entity takes over a performing of the corresponding case-insensitive FSE operation; and b) performing the corresponding case-insensitive FSE operation by the first computing entity.

Handling case sensitivity of filenames

A method that includes instructing multiple compute entities, by an expansion manager and during a first phase of the stored entity metadata re-balancing, to: lookup any stored entity metadata by using: (a) a current translation function for linking identifiers of stored entities to stored entities metadata; wherein the current translation function is based on a size of a current storage space allocated to stored entities metadata; and (b) a next translation function for linking identifiers of stored entities to stored entities metadata; wherein the next translation function is based on a size of a next storage space allocated to stored entities metadata; wherein the current space is expanded during the expansion of the storage system to provide the next storage space, and to update any stored entity metadata accessed using the current translation function without updating stored entity metadata accessed using the next translation function.

Stored entity metadata re-balancing upon storage expansion

A method for lock metadata re-balancing upon an expansion of a storage system, the method may include instructing multiple compute entities, by an expansion manager and during a first phase of the lock metadata re-balancing, to acquire locks by using: (a) a current translation function for linking addresses of current lock metadata units (LMUs) to identifiers of stored entities; wherein the current translation function is based on a size of a current storage space allocated to the lock metadata; and (b) a next translation function for linking addresses of new LMUs to identifiers of stored entities; wherein the new translation function is based on a size of a new storage space allocated to the lock metadata; wherein the current space is expanded during the expansion of the storage system to provide the next storage space. Instructing the multiple compute entities, by the expansion manager, and during another phase of the lock metadata re-balancing that follows a release of all current LMUs, to acquire locks and check status of locks using the next translation function without using the current translation function.

Lock metadata re-balancing upon storage expansion

A method for hierarchical workload allocation in a storage system, the method may include determining to reallocate a compute workload of a current compute core of the storage system; wherein the current compute core is responsible for executing a workload allocation unit that comprises one or more first type shards; and reallocating the compute workload by (a) maintaining the responsibility of the current compute core for executing the workload allocation unit, and (b) reallocating at least one first type shard of the one or more first type shards to a new workload allocation unit that is allocated to a new compute core of new compute cores.

Hierarchical workload allocation in a storage system

A method for preventing file system case related errors, the method may include receiving, by a storage system, an indication that a case insensitive file system client intends to cache a first file of a file system; searching for match between (a) at least a part of a case-insensitive version of a case-sensitive pathname of the first file, and (b) at least a part of a case-insensitive version of a case-sensitive pathname of a second file that belongs to the file system and differs from the first file; and preventing a caching of the first file by the case insensitive file system client.

Managing case sensitivity in a multi file protocol environment

A snapshot of selected objects in a source repository is created in response to the user-initiated replication. The snapshot is designated as a snapshot replication job. Unsynchronized objects in a target destination are detected by comparing a state of the selected objects in the snapshot with a current state of the target destination at the time of execution of the snapshot replication job. An object identification (ID) and a state ID are sent to the target destination and used to locate and compare the selected object with a current state of the selected object in the target destination.

User initiated replication in a synchronized object replication system

For efficient centralized stream initiation and retry control in a computing environment, using a centralized data streams management module for both managing when data streams should be opened and sent from a source location to a destination and for determining when to reattempt opening data streams sent from the source location to the destination after an nth number of consecutive failed attempts using an incrementing time calculation. The incrementing time calculation computes a dynamically calculated time period.

Efficient centralized stream initiation and retry control

For increasing data storage capacity using a processor device, increasing logical capacity of data storage having a multiplicity of storage units containing fixed partitioned segments by moving a selected number of the fixed partitioned segments in one or more storage units to a reserved storage unit.

Increasing data storage capacity

A snapshot of selected objects in a source repository is created in response to the user-initiated replication. The snapshot is designated as a snapshot replication job. Unsynchronized objects in a target destination are detected by comparing a state of the selected objects in the snapshot with a current state of the target destination at the time of execution of the snapshot replication job. One of the selected objects is retrieved in an ordered manner from the snapshot. An object identification (ID) and a state ID are sent to the target destination and used to locate and compare the selected object with a current state of the selected object in the target destination.

A unique file-system node identification (ID) is created for each newly created node in a file system repository by combining a grid identification (ID), a repository identification (ID), and a node identification (ID) to form the unique file-system node ID. The unique file-system node ID is associated with a unique association identification (ID) thereby linking the node from a source repository to a target repository to form an association when performing a replication operation between the source repository and the target repository. Upon removing the association, the unique association ID is unlinked from a replication policy in a source repository, where an initiated replication operation is disabled.

Replica identification and collision avoidance in file system replication

A plurality of server processor workers is configured for the distributed parallel processing of deduplicated data entities in a plurality of chunks. The deduplicated data processing rate is regulated using a rate control mechanism. The rate control mechanism incorporates a debt/credit algorithm specifying which of the plurality of workers processing the deduplicated data entities must wait for each of a plurality of calculated required sleep times, the calculated required sleep times being calculated as a best fit between a maximum allowable sleep time, a delta vector, and a limit per time vector, wherein the required sleep time is one of the plurality of calculated required sleep times.

Deduplicated data processing rate control

A task is marked as dependent upon a preceding task. The task that is attempted to be taken for execution from a head of a pending task queue that is marked is deferred. The deferred task is removed from the pending task queue and placed in a deferred task queue. The deferred task is reinserted back into the pending task queue for execution upon determining that the preceding tasks are completed.

Prioritizing deferred tasks in pending task queue based on creation timestamp

In one embodiment, by way of example only, a resource threshold limits the resource usage of each of the multiplicity of clients. The resource threshold of each of the multiplicity of clients increases over time and decreases when the resources deplete and a resource request of a client is rejected. A subset of clients is created from the multiplicity of clients having a resource usage greater than the resource usage of the rejected client. A dynamic average of a normalized number of resources is calculated from the subset of clients. The resource threshold of each client from the subset of clients is decreased based on the dynamic average of the subset of clients, and further acquisition of resources is restricted from each client from the subset of clients in order to bring the resource usage of each of the subset of clients under the resource threshold.

Growing community of inventors

Omer, Israel

Oded Sonin