Abstract: A mechanism for selectively disabling and enabling read caching based on past performance of the cache and current read/write requests. The system improves overall performance by using an autonomic algorithm to disable read caching for regions of backend disk storage (i.e., the backstore) that have had historically low cache hit ratios. The result is that more cache becomes available for workloads with larger hit ratios, and less time and machine cycles are spent searching the cache for data that is unlikely to be there.

Abstract: An apparatus, system, and method are disclosed for parallelizing access to shared assets. A size module determines if a size of a first spoke of a plurality of spokes is greater than a low threshold. The spokes are arranged in a circular order wherein a global spoke pointer identifies the first spoke as a current spoke. Each spoke is configured as an asset queue. A request module requests an asset from the first spoke if the size of the first spoke is greater than the low threshold. A rotate module rotates the global spoke pointer to a second spoke of the plurality of spokes if the size of the first spoke is not greater than the low threshold, wherein the second spoke becomes the current spoke.

Abstract: An apparatus, system and method to optimize load distribution across logical and physical resources in a storage system. An apparatus in accordance with the invention may include an availability module and an allocation module. The availability module may dynamically assign values to resources in a hierarchical tree structure. Each value may correspond to an availability parameter such as allocated volumes, current resource utilization, and historic resource utilization. The allocation module may serially process the values and allocate a load to a least busy resource in the hierarchical tree structure based on the assigned values.

Abstract: A method, system and computer program product for minimizing read response time in a storage subsystem including a plurality of resources is provided. A middle logical block address (LBA) is calculated for a read request. A preferred resource of the plurality of resources is determined by calculating a minimum seek time based on a closest position to a last position of a head at each resource of the plurality of resources, estimated from the middle LBA. The read request is directed to at least one of the preferred resource or an alternative resource.

Abstract: A method, system and computer program product for minimizing read response time in a storage subsystem including a plurality of resources is provided. A middle logical block address (LBA) is calculated for a read request. A preferred resource of the plurality of resources is determined by calculating a minimum seek time based on a closest position to a last position of a head at each resource of the plurality of resources, estimated from the middle LBA. The read request is directed to at least one of the preferred resource or an alternative resource.

Abstract: An apparatus, system and method to optimize load distribution across logical and physical resources in a storage system. An apparatus in accordance with the invention may include an availability module and an allocation module. The availability module may dynamically assign values to resources in a hierarchical tree structure. Each value may correspond to an availability parameter such as allocated volumes, current resource utilization, and historic resource utilization. The allocation module may serially process the values and allocate a load to a least busy resource in the hierarchical tree structure based on the assigned values.

Abstract: An apparatus, system, and method are disclosed for parallelizing access to shared assets. A size module determines if a size of a first spoke of a plurality of spokes is greater than a low threshold. The spokes are arranged in a circular order wherein a global spoke pointer identifies the first spoke as a current spoke. Each spoke is configured as an asset queue. A request module requests an asset from the first spoke if the size of the first spoke is greater than the low threshold. A rotate module rotates the global spoke pointer to a second spoke of the plurality of spokes if the size of the first spoke is not greater than the low threshold, wherein the second spoke becomes the current spoke.