Abstract: Embodiments of the present invention relate to dynamic feedback-based throughput control for storage systems. In one embodiment, a method of and computer program product for storage throughput control are provided. A plurality of I/O requests is received at a rate controller. The plurality of I/O requests is sent from the rate controller to a storage system at a first rate. Throughput of the storage system is observed. The first rate is dynamically adjusted based on the variance between the observed throughput of the storage system and the first rate.

Abstract: Embodiments of the invention relate to synchronization of data in a shared pool of configurable computer resources. An image of the filesystem changes, including data and metadata, is captured in the form of a consistency point. Sequential consistency points are created, with changes to data and metadata in the filesystem between sequential consistency captured and placed in a queue for communication to a target filesystem at a target site. The changes are communicated as a filesystem operation, with the communication limited to the changes captured and reflected in the consistency point.

Abstract: Embodiments of the invention relate to dynamic application migration in a shared pool of configurable computer resources with disaster recovery support. Data from an application is replicated from local data storage to remote data storage. A consistency point of the data is created in both the local data storage and the remote data storage. The application may be migrated to a second data site with separate local data storage. The migration may be planned or unplanned. Based upon the created consistency point, a consistent set of application data may be requested to support a read operation from the migrated application.

Abstract: A system facilitates access to data in a network. One implementations includes a caching layer function that is configured to integrate into a local cluster file system, to cache local file data in a cache based on fetching file data on demand from a remote cluster file system, and to operate on a multi-node cache cluster.

Abstract: Embodiments relate to supporting transaction data committed to a stable storage. Committed data in the cluster is stored in the persistent cache layer and replicated and stored in the cache layer of one or more secondary nodes. One copy is designated as a master copy and all other copies are designated as replica, with an exclusive write lock assigned to the master and a shared write lock extended to the replica. An acknowledgement of receiving the data is communicated following confirmation that the data has been replicated to each node designated to receive the replica. Managers and a director are provided to support management of the master copy and the replicas within the file system, including invalidation of replicas, fault tolerance associated with failure of a node holding a master copy, recovery from a failed node, recovered of the file system from a power failure, and transferring master and replica copies within the file system.

Abstract: Aspects of the invention are provided to support partial file caching on a file system block boundary. All read requests are converted so that offset and count are aligned on a block boundary. Data associated with read requests is first satisfied from local cache, with cache misses supported with a call to persistent or remote system. Similarly, for a write request, any partial blocks are aligned to the block boundary. Data associated with the write request is performed on local cache and placed in a queue for replay to the persistent or remote system.

Abstract: Scalable caching of remote file data in cluster file systems is provided. One implementation involves maintaining a cache in a local cluster file system and caching local file data in the cache by fetching file data on demand from the remote cluster file system into the local cached file system over the network. The local file data and metadata corresponds to the remote file data and metadata in the remote cluster file system. Updates made to the local file data and metadata are pushed back to the remote cluster file system asynchronously.

Abstract: Embodiments of the present invention relate to dynamic feedback-based throughput control for storage systems. In one embodiment, a method of and computer program product for storage throughput control are provided. A plurality of I/O requests is received at a rate controller. The plurality of I/O requests is sent from the rate controller to a storage system at a first rate. Throughput of the storage system is observed. The first rate is dynamically adjusted based on the variance between the observed throughput of the storage system and the first rate.

Abstract: Scalable caching of remote file data in cluster file systems is provided. One implementation involves maintaining a cache in a local cluster file system and caching local file data in the cache by fetching file data on demand from the remote cluster file system into the local cached file system over the network. The local file data and metadata corresponds to the remote file data and metadata in the remote cluster file system. Updates made to the local file data and metadata are pushed back to the remote cluster file system asynchronously.

Abstract: Embodiments of the invention relate to supporting transaction data committed to a stable storage. Committed data in the cluster is stored in the persistent cache layer and replicated and stored in the cache layer of one or more secondary nodes. One copy is designated as a master copy and all other copies are designated as replica, with an exclusive write lock assigned to the master and a shared write lock extended to the replica. An acknowledgement of receiving the data is communicated following confirmation that the data has been replicated to each node designated to receive the replica. Managers and a director are provided to support management of the master copy and the replicas within the file system, including invalidation of replicas, fault tolerance associated with failure of a node holding a master copy, recovery from a failed node, recovered of the file system from a power failure, and transferring master and replica copies within the file system.

Abstract: Embodiments of the invention relate to efficiently employing checksums for shared nothing clustered filesystems. Tools are provided to compute the checksum in response to a read transaction and to utilize the computed checksum to prevent serving corrupted data. Multiple levels of data replication are provided. The checksum computation functions within the multiple levels and addresses a specified data block that is the subject of the read transaction.

Abstract: Aspects of the invention are provided to support partial file caching on a file system block boundary. All read requests are converted so that offset and count are aligned on a block boundary. Data associated with read requests is first satisfied from local cache, with cache misses supported with a call to persistent or remote system. Similarly, for a write request, any partial blocks are aligned to the block boundary. Data associated with the write request is performed on local cache and placed in a queue for replay to the persistent or remote system.

Abstract: A method for accessing data stored in a distributed caching storage system containing a home cluster and a secondary cluster is provided. A first copy of a file is stored on the home cluster and a second copy of the file is stored on the secondary cluster. The second copy of the file is associated with an inode data structure having a consistency attribute. An input/output request is received directed to the file and indicates that file is in an inconsistent state by updating the inode's consistency attribute. The first copy and the second copy of the file is updated according to the received input/output request and it is determined whether the first copy and the second copy were updated successfully. The maintaining of the inode's consistency attribute is indicative of the inconsistent state of the file.

Abstract: Embodiments of the present invention relate to dynamic feedback-based throughput control for storage systems. In one embodiment, a method of and computer program product for storage throughput control are provided. A plurality of I/O requests is received at a rate controller. The plurality of I/O requests is sent from the rate controller to a storage system at a first rate. Throughput of the storage system is observed. The first rate is dynamically adjusted based on the variance between the observed throughput of the storage system and the first rate.

Abstract: Embodiments of the invention relate to optimizing a file system for different types of applications in a compute cluster using dynamic block size granularity. An exemplary embodiment includes reserving a predetermined number of storage allocation regions for each node in a cluster, wherein each storage allocation region comprises a set of contiguous data blocks on a single storage disk of locally attached storage; using a contiguous set of data blocks on a single storage disk of locally attached storage as a file system's block allocation granularity for striping data to be stored in the file system for a compute operation in the cluster; and using a single data block of a shared storage subsystem as the file system's block allocation granularity for striping data to be stored in the file system for a data access operation in the cluster.

Abstract: Aspects of the invention are provided to support partial file caching on a file system block boundary. All read requests are converted so that offset and count are aligned on a block boundary. Data associated with read requests is first satisfied from local cache, with cache misses supported with a call to persistent or remote system. Similarly, for a write request, any partial blocks are aligned to the block boundary. Data associated with the write request is performed on local cache and placed in a queue for replay to the persistent or remote system.

Abstract: Embodiments of the invention relate to creating and maintaining consistent data sets in a shared pool of configurable computer resources to support disaster recovery support. Data from an application is stored in local data storage and replicated to another data storage. A consistency point of the data is created in both of the data storage, with the consistency point representing an identical data set at a point-in-time. Based upon the created consistency points, a consistent set of application data may be requested to support a read operation for a migrated application.

Abstract: Embodiments of the invention relate to creating and maintaining consistent data sets in a shared pool of configurable computer resources to support disaster recovery support. Data from an application is stored in local data storage and replicated to another data storage. A consistency point of the data is created in both of the data storage, with the consistency point representing an identical data set at a point-in-time. Based upon the created consistency points, a consistent set of application data may be requested to support a read operation for a migrated application.

Abstract: An adaptive data replication and caching system configured for deployment at the network edge. This system can include a data caching system configured to cache application data in a data store and to satisfy queries for application data from the cache. The system also can include a data replication system configured to replicate application data in the data store and to satisfy queries for application data against the replicated application data. Finally, the adaptive data replication and caching system can include a configurator for configuring portions of the data store for exclusive use by the data caching system and the data replication system.

Abstract: A method for accessing data stored in a distributed caching storage system containing a home cluster and a secondary cluster is provided. A first copy of a file is stored on the home cluster and a second copy of the file is stored on the secondary cluster. The second copy of the file is associated with an inode data structure having a consistency attribute. An input/output request is received directed to the file and indicates that file is in an inconsistent state by updating the inode's consistency attribute. The first copy and the second copy of the file is updated according to the received input/output request and it is determined whether the first copy and the second copy were updated successfully. The maintaining of the inode's consistency attribute is indicative of the inconsistent state of the file.