A method, hybrid server system, and computer program product, prefetch data. A set of prefetch requests associated with one or more given datasets residing on the server system are received from a set of accelerator systems. A set of data is prefetched from a memory system residing at the server system for at least one prefetch request in the set of prefetch requests. The set of data satisfies the at least one prefetch request. The set of data that has been prefetched is sent to at least one accelerator system, in the set of accelerator systems, associated with the at least one prefetch request.

Once or more embodiments manage access to data by accelerator systems in an out-of-core processing environment. In one embodiment, a request from an accelerator system is received for access to a given data set. An access context associated with the given data set is determined The accelerator system is dynamically configured, based on the access context that has been determined, based on the access context that has been determined, to one of access the given data set directly from the server system; locally store a portion of the given data set in a memory; and locally store all of the given data set in the memory.

A method, hybrid server system, and computer program product, prefetch data. A set of prefetch requests associated with one or more given datasets residing on the server system are received from a set of accelerator systems. A set of data is prefetched from a memory system residing at the server system for at least one prefetch request in the set of prefetch requests. The set of data satisfies the at least one prefetch request. The set of data that has been prefetched is sent to at least one accelerator system, in the set of accelerator systems, associated with the at least one prefetch request.

Techniques for distributed cache management are provided. A server having backend resource includes a global cache and a global cache agent. Individual clients each have client cache agents and client caches. When data items associated with the backend resources are added, modified, or deleted in the client caches, the client cache agents report the changes to the global cache agent. The global cache agent records the changes and notifies the other client cache agents to update a status of the changes within their client caches. When the changes are committed to the backend resource each of the statuses in each of the caches are updated accordingly.

Disclosed herein are methods, systems, and processes to provide coherency across disjoint caches in clustered environments. It is determined whether a data object is owned by an owner node, where the owner node is one of multiple nodes of a cluster. If the owner node for the data object is identified by the determining, a request is sent to the owner node for the data object. However, if the owner node for the data object is not identified by the determining, selects a node in the cluster is selected as the owner node, and the request for the data object is sent to the owner node.

Once or more embodiments manage access to data by accelerator systems in an out-of-core processing environment. In one embodiment, a request from an accelerator system is received for access to a given data set. An access context associated with the given data set is determined. The accelerator system is dynamically configured, based on the access context that has been determined, based on the access context that has been determined, to one of access the given data set directly from the server system; locally store a portion of the given data set in a memory; and locally store all of the given data set in the memory.

A method, hybrid server system, and computer program product, prefetch data. A set of prefetch requests associated with one or more given datasets residing on the server system are received from a set of accelerator systems. A set of data is prefetched from a memory system residing at the server system for at least one prefetch request in the set of prefetch requests. The set of data satisfies the at least one prefetch request. The set of data that has been prefetched is sent to at least one accelerator system, in the set of accelerator systems, associated with the at least one prefetch request.

Implementations described and claimed herein provide a method and system for managing execution of commands for a storage device, the method comprising identifying individual streams processing read ahead operations in a storage controller, determining an amount of read ahead data that each individual stream is processing in the read ahead operations, determining a total amount of read cache available for the storage controller, and determining a total amount of read ahead data that all the individual streams are processing in the read ahead operations.

A storage control apparatus is configured to: detect a ratio of a number of first data blocks stored by a first process including executing deduplication to a number of second data blocks stored by a second process not including executing the deduplication in data blocks stored in a storage device, and determine which of the first and second processes to use to execute a write process for a third write data block which is newly requested to be written so that the ratio approaches a target ratio based on a load of a third process for the second data blocks and a lower limit target value of a number of write requests processable per unit time in response to write requests, the third process including executing the deduplication for each of the second data blocks and storing again the second data block in any storage device.

A file storage system for storing data of a file received from a client includes a back-end file server in which the data is stored. The system includes a cache appliance in communication with the file server, such that the appliance stores portions of the data or attributes of the file, and uses the stored data or attributes to process file system requests received from the client, and which reads and writes data and attributes to the back-end file server independently. A system for responding to a file system protocol request in regard to a back-end server includes a token server. The system includes a plurality of cache appliances in communication with the token server, each of which receives tokens from the token server to synchronize access to data and attributes caches of the cache appliances, and reading and writing data and attributes to the back-end servers when tokens are revoked, the cache appliance having persistent storage in which data are stored, and the token server having persistent storage in which tokens are stored. A storage system includes a plurality of backend servers. The system includes a token server which grants permission to read and write file attributes and data system, and includes a plurality of cache appliances in communication with at least one of the backend servers and the token server for processing an incoming NFS request to the one backend server. Each cache appliance comprises an NFS server which converts incoming NFS requests into cache manager operations; a token client module in communication with the token server having a cache of tokens obtained from the token server; a cache manager that caches data and attributes and uses tokens from the token client module to ensure that the cached data or attributes are the most recent data or attributes, and an NFS client which sends outgoing NFS requests to the back-end file server. Methods for storing data of a file received from a client.