Red Hat, Inc.

Emc Corporation

Jeffrey Jon Darcy

Shyamsundar Ranganathan

Anand Vishweswaran Avati

Pranith Kumar Karampuri

Raghavendra Gowdappa

Xiaoye Jiang

Uresh K Vahalia

Robert J Relyea

Boris Zuckerman

David Lutterkort

Vijay Bellur

Ronald Curtis Searls

Efficient authentication in a file system with multiple security groups is disclosed. A file system (FS) executing on at least one processor device receives, from a first client application of a plurality of client applications, a request to access a first object, the request including a unique object ID that identifies the first object. The FS determines, based on a data structure maintained by the FS and inaccessible to the first client application, that the first client application is associated with a first security group of a plurality of different security groups. The FS determines, based on metadata of the first object, that the first object is associated with the first security group, and grants the first client application access to the first object.

Efficient authentication in a file system with multiple security groups

Systems and methods for enforcing consistency of file to storage node mappings in distributed file systems. An example method may include: receiving a request to perform a file operation with respect to a file associated with a directory of a distributed file system; identifying a directory layout version number associated with the directory; identifying a volume layout version number associated with a volume in which the directory resides; and responsive to determining that the directory layout version number matches the volume layout version number, performing the requested file operation with respect to the file.

Enforcing consistency of file to storage node mapping

A first storage server of the file system receives a request to perform an operation on the data file. The operation is to be performed on a first replica of the data file stored at the first storage server and a second replica of the data file stored at a second storage server of the file system. The first storage server configures first metadata associated with a first index file to indicate that the operation is to be performed on a first portion of the first replica. The first storage server determines the second replica is an outdated state indicating that the operation on the second replica has not been performed by the second storage server. In response to the second replica being in the outdated state, updating a first portion of the second replica identified in view of the first metadata and corresponding to the first portion of the first replica.

Granular data self-healing

A caseless file lookup in a distributed file system is disclosed. A request to store a file in a volume may be received. In response to the request, the type of client system issuing the request may be identified. Based on the type of client system that has been identified, a first hash value or a second hash value may be calculated for the name of the file to generate and assign a hash value to the file. The file may be stored in a storage resource based on the first or second hash value that has been generated for the file.

Caseless file lookup in a distributed file system

A method and system for storing an object in a distributed storage system are provided. An example system includes an object name identifier that receives a request to save an object having an initial object name within a file system space and determines, based on a set of regular expressions, whether the initial object name is a temporary object name. The example system also includes a converter that, when the initial object name is determined to be a temporary object name, transforms, based on the set of regular expressions, the initial object name into a permanent object name. The example system further includes a selector that applies a function to the permanent object name and selects, based on a result of the function applied to the permanent object name, a server of a plurality of servers. The object is sent to the selected server for storage.

Storing an object in a distributed storage system

A request to store a file in a directory of a volume associated with a distributed file system may be received. A server of a plurality of servers of the distributed file system that is associated with the file may be identified. A first version number associated with the directory and a second version number associated with the volume may be compared. The file may be stored in the directory at the identified server when the first version number matches the second version number and when the file is not currently stored at the server.

File lookup in a distributed file system

Systems and methods for allocating file system metadata to storage nodes of a distributed file system. An example method may include: defining, by a processing device, a plurality of tokens, wherein each token comprises a sequence of bits; associating each token of the plurality of tokens with a metadata node of a cluster of metadata nodes; receiving a request to create a directory; selecting, among the plurality of tokens, a token associated with the directory; generating a directory identifier comprising the selected token; and creating, on a metadata node associated with the selected token, a directory record comprising the directory identifier.

Allocating file system metadata to storage nodes of distributed file system

Replication in a distributed file system is described. A processing device determines that a completion threshold has been satisfied in view of at least a received indication of completion of a new data modification request, wherein the completion threshold comprises a number of server devices of the distributed file system for which the new data modification has been performed thereon. The processing device provides, in response to determining that the completion threshold has been satisfied, a message to a client device that the new data modification request was performed successfully.

Replication with adjustable consistency levels

A file may be identified and a plurality of rules associated with a plurality of storage servers may be received. An order for the rules may be identified. A particular storage server of the plurality of storage servers that satisfies a condition for each of the rules in view of the order for the rules and a characteristic associated with each of the plurality of storage servers. The file may be stored at the particular storage server that satisfies the condition for each of the rules.

File classification in a distributed file system

Systems and methods for enforcing consistency of file to storage node mappings in distributed file systems. An example method may include: receiving a request to perform a file operation with respect to a file associated with a volume of a distributed file system; identifying, among a plurality of storage nodes of the distributed file system, a storage node associated with the file; obtaining a lock of a layout of the volume; and responsive to determining, by a processing device, that a directory layout version number matches a volume layout version number, performing the requested file operation with respect to the file.

Embodiments relate to systems and methods for replicating a group of data objects within a storage network. In particular, a module can designate an object in the storage network as an anchor object. The module can further evaluate a replication policy of the anchor object to determine a destination resource on which to replicate a related group of data objects. Further, the module can located the related group of data objects, and define a replication policy on each of the data objects that specifies that each of the data objects follow the anchor object. The module can replicate each of the data objects as well as the anchor object on the destination resource.

Replicating a group of data objects within a storage network

Replication in a distributed file system is described. A processing device receives a new data modification request from a client device. The processing device provides the request in a sequence to a server. The server is to perform the new data modification request in view of the sequence. The processing device receives an indication of completion of the new data modification request from the server. The processing device determines that a completion threshold has been satisfied in view of the received indication of completion of the new data modification request. The processing device provides, in view of the determining, a message to the client device that the new data modification request was performed successfully.

Replication with adustable consistency levels

A processing device identifies a file (or data block) and selects a storage location for the file based on a plurality of data distribution policies. To select the storage location, the processing device applies, at a top distribution layer, a first data distribution policy to select a first storage pool from a plurality of storage pools based on a first storage criterion. The processing device then applies, at a bottom distribution layer, a second data distribution policy associated with the first storage pool to select a first storage node from a plurality of storage nodes in the first storage pool based on a second storage criterion. The processing device stores at least a portion of the file in the first storage node of the first storage pool.

Tiered distributed storage policies

Data connections from a remote login session are identified and an identification query may be transmitted by the remote login session to endpoints (e.g., servers) of the data connections. In response to the identification query, identification information may be received from the endpoints. Furthermore, the identification information may be compared with a local identification of a local server that hosts the remote login session. A file created by the remote login session may be stored at an endpoint over a data connection where the local identification matches a portion or all of the identification information of the endpoint of the data connection.

Growing community of inventors

Lexington, MA, United States of America

Jeffrey Jon Darcy