Abstract: A first cache of a first IOA is detected storing an amount of data that satisfies a memory shortage threshold. A request for extra memory for the first IOA is transmitted. The request is sent in response to detecting that the first cache stores the amount of data that satisfies the memory shortage threshold. The request is transmitted to a plurality of IOAs of a computer system. A second cache of a second IOA is detected storing an amount of data that satisfies a memory dissemination threshold. Memory of the second cache is allocated to the first cache. The memory is allocated in response to the request and the amount of data in the second cache satisfying the memory dissemination threshold.

Abstract: A first cache of a first IOA is detected storing an amount of data that satisfies a memory shortage threshold. A request for extra memory for the first IOA is transmitted. The request is sent in response to detecting that the first cache stores the amount of data that satisfies the memory shortage threshold. The request is transmitted to a plurality of IOAs of a computer system. A second cache of a second IOA is detected storing an amount of data that satisfies a memory dissemination threshold. Memory of the second cache is allocated to the first cache. The memory is allocated in response to the request and the amount of data in the second cache satisfying the memory dissemination threshold.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization model processing in Coherent Accelerator Processor Interface (CAPI) adapters. The CAPI adapter includes an authorization table and a file system authorization function to authenticate data access for a client at an extent granularity and to prevent an application from accessing unauthorized data in the CAPI adapter. Each authorization table entry provides for the CAPI client, a CAPI client identification (ID), a CAPI server register space assigning resource ownership to the CAPI client with a CAPI set of allowed functions.

Abstract: Disclosed are embodiments for supporting dynamic tier remapping of data stored in a hybrid storage system. One embodiment includes a storage controller and firmware, where the firmware maintains a plurality of mapping elements, where each mapping element includes a plurality of group identifiers, where each group identifier is configured to indicate a mapping of a logical block addresses, and where the storage controller performs: receiving a read command including a logical block address; parsing the logical block address to determine a mapping element and a group identifier; determining, for a particular mapping element of the plurality of elements, whether the particular mapping element is locked, wherein the particular mapping element corresponds to the mapping element of the logical block address; and dependent upon the particular mapping element, queuing the read command for firmware processing or remapping the logical block address.

Abstract: A method and controller for implementing storage adapter enhanced write cache management, and a design structure on which the subject controller circuit resides are provided. The controller includes a hardware write cache engine implementing hardware acceleration for storage write cache management. The hardware write cache engine monitors cache levels used for managing cache destage rates and thresholds for destages from storage write cache substantially without using firmware for greatly enhancing performance.

Abstract: Caching data in a redundant array of independent disks (RAID) storage system including receiving an operation instruction targeting a location in an attached memory of the RAID storage system, wherein the attached memory temporarily stores data for storage on RAID storage devices, and wherein the operation instruction is one selected from a group consisting of a read instruction and a write instruction; redirecting, based on a content of the operation instruction, the operation instruction from the attached memory to the embedded memory on the RAID storage system; and servicing the operation instruction by accessing a portion of the embedded memory corresponding to the location in the attached memory of the RAID storage system.

Abstract: Caching data in a redundant array of independent disks (RAID) storage system including receiving an operation instruction targeting a location in an attached memory of the RAID storage system, wherein the attached memory temporarily stores data for storage on RAID storage devices, and wherein the operation instruction is one selected from a group consisting of a read instruction and a write instruction; redirecting, based on a content of the operation instruction, the operation instruction from the attached memory to the embedded memory on the RAID storage system; and servicing the operation instruction by accessing a portion of the embedded memory corresponding to the location in the attached memory of the RAID storage system.

Abstract: Disclosed are embodiments for supporting dynamic tier remapping of data stored in a hybrid storage system. One embodiment includes a storage controller and firmware, where the firmware maintains a plurality of mapping elements, where each mapping element includes a plurality of group identifiers, where each group identifier is configured to indicate a mapping of a logical block addresses, and where the storage controller performs: receiving a read command including a logical block address; parsing the logical block address to determine a mapping element and a group identifier; determining, for a particular mapping element of the plurality of elements, whether the particular mapping element is locked, wherein the particular mapping element corresponds to the mapping element of the logical block address; and dependent upon the particular mapping element, queuing the read command for firmware processing or remapping the logical block address.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization model processing in Coherent Accelerator Processor Interface (CAPI) adapters. The CAPI adapter includes an authorization table and a file system authorization function to authenticate data access for a client at an extent granularity and to prevent an application from accessing unauthorized data in the CAPI adapter. Each authorization table entry provides for the CAPI client, a CAPI client identification (ID), a CAPI server register space assigning resource ownership to the CAPI client with a CAPI set of allowed functions.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization and deauthorization processing for a Coherent Accelerator Processor Interface (CAPI) adapter. The Application Client, such as an Application Child Client sends a Delete Authorizations command to the CAPI Adapter via the Client CAPI Server Registers assigned to the specific Application Client. The CAPI Adapter deletes the Authorizations in all Lists in the Delete Authorizations command.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization processing for a Coherent Accelerator Processor Interface (CAPI) adapter. An Application Client requests authorization to a File from a system processor file system. The file system validates the request, determines the location of each Extent that comprises the File, and requests authorization to each Extent from a System CAPI Authorization manager. The System CAPI Authorization manager requests the CAPI Client manager to assign a Child Client ID and CAPI Server Register range to the requesting Application Client and requests a previously authorized CAPI Parent Client to authorize the Child ID to the list of Extents. The CAPI Parent Client sends a Create Authorizations command to the CAPI Adapter via the Parent's CAPI Server Registers.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization command flow processing for a Coherent Accelerator Processor Interface (CAPI) adapter. An Application Client builds a command including start LBA and number of LBAs and Child Authorization Handle. The Application Client sends the command directly to the CAPI Adapter via the Application Clients CAPI Server Registers assigned to the specific Application Client. The CAPI adapter validate that the requesting Client is authorized to perform the command using the Authorization Handle and the receiving CAPI Server Register address. The CAPI Adapter executes the validated command and sends completion back to the Application Client.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization processing for a Coherent Accelerator Processor Interface (CAPI) adapter. An Application Client requests authorization to a File from a system processor file system. The file system validates the request, determines the location of each Extent that comprises the File, and requests authorization to each Extent from a System CAPI Authorization manager. The System CAPI Authorization manager requests the CAPI Client manager to assign a Child Client ID and CAPI Server Register range to the requesting Application Client and requests a previously authorized CAPI Parent Client to authorize the Child ID to the list of Extents. The CAPI Parent Client sends a Create Authorizations command to the CAPI Adapter via the Parent's CAPI Server Registers.

Abstract: A method for operating a RAID storage system includes configuring the RAID storage devices to receive in a read or write command a byte count, receiving a first data block to write to the storage system, compressing the received first data block to generate a first compressed data block, and then storing the first compressed data block memory. The method additionally includes executing a set of RAID operations to perform a partial stripe update, including: retrieving a second compressed data block from memory; determining a physical size of the second compressed data block; generating, based on the second compressed data block and the physical size, redundant data corresponding with the second compressed data block; and writing the second compressed data block and the redundant data by transmitting a write command including the second compressed data block, the redundant data, and the physical size to the set of RAID storage devices.

Abstract: A method, system and computer program product are provided for implementing block extent granularity authorization command flow processing for a Coherent Accelerator Processor Interface (CAPI) adapter. An Application Client builds a command including start LBA and number of LBAs and Child Authorization Handle. The Application Client sends the command directly to the CAPI Adapter via the Application Clients CAPI Server Registers assigned to the specific Application Client. The CAPI adapter validate that the requesting Client is authorized to perform the command using the Authorization Handle and the receiving CAPI Server Register address. The CAPI Adapter executes the validated command and sends completion back to the Application Client.

Abstract: Embodiments described herein include a PSL engine that includes various memory elements that permit the engine to grant locks on particular portions of data in a stripe in a RAID storage system. The PSL engine can assign (or lock) different blocks of the stripe for different operations. The PSL engine can grant locks to multiple operations for the same stripe if the operations access mutually exclusive blocks of the stripe. Each time a new operation is requested, the PSL engine determines whether the operation would affect a stripe data block that is currently assigned to another operation. If the new operation corresponds to a block of data in the stripe that includes data locked by another operation, the PSL engine assigns the new operation to a wait list. In one embodiment, the PSL engine maintains a wait list for each of the stripes in the RAID system.

Abstract: Operating a RAID array with unequal stripes, the RAID array comprising N number of drives, where each RAID stripe includes P number of parity drives and N-P number of data drives, including buffering, by a RAID controller, write operations received from a host, each write operation specifying data to be written to the RAID array; distributing, by the RAID controller, the data to be written amongst N-P write groups, including: dividing the data into chunks of a sub-stripe size, wherein the sub-stripe size is less than a parity stripe size; and assigning the chunks, in round-robin order, to the N-P write groups; calculating parity from the N-P write groups; and writing the N-P write groups and the calculated parity as a first RAID stripe to the RAID array.

Abstract: A method and controller for implementing storage adapter enhanced write cache management, and a design structure on which the subject controller circuit resides are provided. The controller includes a hardware write cache engine implementing hardware acceleration for storage write cache management. The hardware write cache engine performs destage operations from storage write cache with minimal firmware involvement to enhance performance.

Abstract: A method and controller for implementing storage adapter enhanced write cache management, and a design structure on which the subject controller circuit resides are provided. The controller includes a hardware write cache engine implementing hardware acceleration for storage write cache management. The hardware write cache engine manages merging data with existing data on fast writes to storage write cache substantially without using firmware for greatly enhancing performance.

Abstract: A method and controller for implementing storage adapter enhanced write cache management, and a design structure on which the subject controller circuit resides are provided. The controller includes a hardware write cache engine implementing hardware acceleration for storage write cache management. The hardware write cache engine performs data age identification in storage write cache substantially without firmware involvement for greatly enhancing performance.