Abstract: Provided are a computer program product, system, and method for determining sectors of a track to stage into cache by training a machine learning module. A machine learning module that receives as input performance attributes of system components affected by staging tracks from the storage to the cache and outputs a staging strategy comprising one of a plurality of staging strategy indicating at least one of a plurality of sectors of a track to stage into the cache. A margin of error is determined based on a current value of a performance attribute and a threshold of the performance attribute. An adjusted staging strategy is determined based on the margin of error. The machine learning module is retrained with current performance attributes to output the adjusted staging strategy.

Abstract: A method for improving asynchronous data replication between a primary storage system and a secondary storage system is disclosed. In one embodiment, such a method includes monitoring, in a cache of the primary storage system, unmirrored data elements needing to be mirrored, but that have not yet been mirrored, from the primary storage system to the secondary storage system. The method maintains a regular LRU list designating an order in which data elements are demoted from the cache. The method determines whether a data element at an LRU end of the regular LRU list is an unmirrored data element. In the event the data element at the LRU end of the regular LRU list is an unmirrored data element, the method moves the data element to a transfer-pending LRU list dedicated to unmirrored data elements in the cache. A corresponding system and computer program product are also disclosed.

Abstract: A method for demoting a selected storage element from a cache memory includes storing favored and non-favored storage elements within a higher performance portion and lower performance portion of the cache memory. The favored storage elements are retained in the cache memory longer than the non-favored storage elements. The method maintains a first favored LRU list and a first non-favored LRU list, associated with the favored and non-favored storage elements stored within the higher performance portion of the cache. The method selects a favored or non-favored storage element to be demoted from the higher performance portion of the cache memory according to life expectancy and residency of the oldest favored and non-favored storage elements in the first LRU lists. The method demotes the selected from the higher performance portion of the cache to the lower performance portion of the cache, or to the data storage devices, according to a cache demotion policy.

Abstract: A first server of a storage controller is configured to communicate with a host via a first bus interface, and a second server of the storage controller is configured to communicate with the host via a second bus interface. Data is written from the host via the first bus interface to a cache of the first server and via the second bus interface to a non-volatile storage of the second server. The data stored in the cache of the first server is periodically compared to the data stored in the non-volatile storage of the second server.

Abstract: A method for dynamically adjusting cache memory partition sizes within a storage system includes computing a read hit ratio for data accessed in each cache partition and an average read hit ratio for all the cache partitions over a time interval. The cache memory includes a higher performance portion (DRAM) and lower performance portion (SCM). The method increases or decreases the partition size for each cache partition by comparing the read hit ratio for the partition to the average read hit ratio for all the partitions. Each cache partition includes maximum and minimum partition sizes, and read hit and read access counters. The SCM portion of the cache memory includes cache partitions reserved for storing data of a specific type, or data used for a specific purpose or with a specific software application. A corresponding storage controller and computer program product are also disclosed.

Abstract: A method for demoting data from a cache comprising heterogeneous memory types is disclosed. The method maintains for a data element in the cache, a write access count that is incremented each time the data element is updated in the cache. The cache includes a higher performance portion and a lower performance portion. The method also maintains, for the data element, a read access count that is incremented each time a data element is read in the cache. The method removes the data element from the higher performance portion of the cache in accordance with a cache demotion algorithm. If the write access count is below a first threshold and the read access count is above a second threshold, the method places the data element in the lower performance portion. A corresponding system and computer program product are also disclosed.

Abstract: A method for demoting a selected storage element from a cache memory includes storing favored and non-favored storage elements within a higher performance portion and lower performance portion of the cache memory. The method maintains a plurality of favored LRU lists and a non-favored LRU list for the higher and lower performance portions of the cache memory. Each favored LRU list contains entries associated with the favored storage elements that have the same unique residency multiplier. The non-favored LRU list includes entries associated with the non-favored storage elements. The method demotes a selected favored or non-favored storage element from the higher and lower performance portions of the cache memory according to a cache demotion policy that provides a preference to favored storage elements over non-favored storage elements based on a computed cache life expectancy, residency time, and the unique residency multiplier.

Abstract: A method for dynamically adjusting cache memory partition sizes within a storage system includes computing a read hit ratio for data accessed in each cache partition and an average read hit ratio for all the cache partitions over a time interval. The cache memory includes a higher performance portion (DRAM) and lower performance portion (SCM). The method increases or decreases the partition size for each cache partition by comparing the read hit ratio for the partition to the average read hit ratio for all the partitions. Each cache partition includes maximum and minimum partition sizes, and read hit and read access counters. The SCM portion of the cache memory includes cache partitions reserved for storing data of a specific type, or data used for a specific purpose or with a specific software application. A corresponding storage controller and computer program product are also disclosed.

Abstract: A method for demoting a selected storage element from a cache memory includes storing favored and non-favored storage elements within a higher performance portion and lower performance portion of the cache memory. The favored storage elements are retained in the cache memory longer than the non-favored storage elements. The method maintains a first favored LRU list and a first non-favored LRU list, associated with the favored and non-favored storage elements stored within the higher performance portion of the cache. The method selects a favored or non-favored storage element to be demoted from the higher performance portion of the cache memory according to life expectancy and residency of the oldest favored and non-favored storage elements in the first LRU lists. The method demotes the selected from the higher performance portion of the cache to the lower performance portion of the cache, or to the data storage devices, according to a cache demotion policy.

Abstract: A method for demoting a selected storage element from a cache memory includes storing favored and non-favored storage elements within a higher performance portion and lower performance portion of the cache memory. The method maintains a plurality of favored LRU lists and a non-favored LRU list for the higher and lower performance portions of the cache memory. Each favored LRU list contains entries associated with the favored storage elements that have the same unique residency multiplier. The non-favored LRU list includes entries associated with the non-favored storage elements. The method demotes a selected favored or non-favored storage element from the higher and lower performance portions of the cache memory according to a cache demotion policy that provides a preference to favored storage elements over non-favored storage elements based on a computed cache life expectancy, residency time, and the unique residency multiplier.

Abstract: Provided are techniques for selecting a disconnect by training a machine learning module. A machine learning module is provided that receives inputs and produces an output. The output produced from the machine learning module based on the inputs for the first I/O operation and an estimated amount of time to acquire resources for a first I/O operation is determined. An actual amount of time to acquire resources for the first I/O operation is determined. The machine learning module is retrained based on the inputs, the output, and the actual amount of time it took to acquire resources for the first I/O operation versus an estimated amount of time to acquire the resources for the first I/O operation. The retrained machine learning module is used to select one of disconnect from a channel, the logical disconnect from the channel, or the physical disconnect from the channel for a second I/O operation.

Abstract: Provided are a computer program product, system, and method for using a track format code in a cache control block for a track in a cache to process read and write requests to the track in the cache. A track format table associates track format codes with track format metadata. A determination is made as to whether the track format table has track format metadata matching track format metadata of a track staged into the cache. A determination is made as to whether a track format code from the track format table for the track format metadata in the track format table matches the track format metadata of the track staged. A cache control block for the track being added to the cache is generated including the determined track format code when the track format table has the matching track format metadata.

Abstract: Provided are a computer program product, system, and method for using a mirroring cache list to mirror modified tracks for a primary storage in a cache to a secondary storage. Indication is made of a modified track for the primary storage stored in the cache in a mirroring cache list. The mirroring cache list is processed to select modified tracks in the cache to transfer to the secondary storage that have not yet been transferred. The selected modified tracks are transferred to the secondary storage. Indication of a modified track is removed from the mirroring cache list in response to demoting the modified track from the cache.

Abstract: Provided are a computer program product, system, and method for using mirroring cache list to demote modified tracks from cache A modified track for a primary storage stored in the cache to mirror to a secondary storage is indicated in a mirroring cache list. The mirroring cache list is processed to select modified tracks in the cache to transfer to the secondary storage that have not yet been transferred. The selected modified tracks in the cache are transferred to the secondary storage. The mirroring cache list is processed to determine modified tracks in the cache to demote from the cache.

Abstract: Provided are a computer program product, system, and method for using a machine learning module to determine an allocation of stage and destage tasks. Storage performance information related to processing of Input/Output (I/O) requests with respect to the storage unit is provided to a machine learning module. The machine learning module receives a computed number of stage tasks and a computed number of destage tasks. A current number of stage tasks allocated to stage tracks from the storage unit to the cache is adjusted based on the computed number of stage tasks. A current number of destage tasks allocated to destage tracks from the cache to the storage unit is adjusted based on the computed number of destage tasks.

Abstract: Provided are a computer program product, system, and method for determining an optimum number of threads to make available per core in a multi-core processor complex to execute tasks. A determination is made of a first processing measurement based on threads executing on the cores of the processor chip, wherein each core includes circuitry to independently execute a plurality of threads. A determination is made of a number of threads to execute on the cores based on the first processing measurement. A determination is made of a second processing measurement based on the threads executing on the cores of the processor chip. A determination is made of an adjustment to the determined number of threads to execute based on the second processing measurement resulting in an adjusted number of threads. The adjusted number of threads on the cores is utilized to execute instructions.

Abstract: Provided are techniques for using real segments and alternate segments in Non-Volatile Storage (NVS). One or more write requests for a track are executed by alternating between storing data in one or more sectors of real segments and one or more sectors of alternate segments for each of the write requests, while setting indicators in a real sector structure and an alternate sector structure. In response to determining that the one or more write requests for the track have completed, the data stored in the one or more sectors of the real segments and in the one or more sectors of the alternate segments are merged to form newly written data. In response to determining that a hardened, previously written data of a track does exist in Non-Volatile Storage (NVS), the newly written data is merged with the hardened, previously written data in the NVS. The merged data is committed.

Abstract: In response to an end of track access for a track in a cache, a determination is made as to whether the track has modified data and whether the track has one or more holes. In response to determining that the track has modified data and the track has one or more holes, an input on a plurality of attributes of a computing environment in which the track is processed is provided to a machine learning module to produce an output value. A determination is made as to whether the output value indicates whether one or more holes are to be filled in the track. In response to determining that the output value indicates that one or more holes are to be filled in the track, the track is staged to the cache from a storage drive.

Abstract: A method for improving asynchronous data replication between a primary storage system and a secondary storage system maintains a cache in the primary storage system. The cache includes a higher performance portion and a lower performance portion. The method monitors, in the cache, unmirrored data elements needing to be mirrored, but that have not yet been mirrored, from the primary storage system to the secondary storage system. The method maintains a regular LRU list designating an order in which data elements are demoted from the cache. The method determines whether a data element at an LRU end of the regular LRU list is an unmirrored data element. In the event the data element at the LRU end is an unmirrored data element, the method moves the data element from the higher performance portion to the lower performance portion. A corresponding system and computer program product are also disclosed.

Abstract: Provided are a computer program product, system, and method for using a machine learning module to select one of multiple cache eviction algorithms to use to evict a track from the cache. A first cache eviction algorithm determines tracks to evict from the cache. A second cache eviction algorithm determines tracks to evict from the cache, wherein the first and second cache eviction algorithms use different eviction schemes. At least one machine learning module is executed to produce output indicating one of the first cache eviction algorithm and the second cache eviction algorithm to use to select a track to evict from the cache. A track is evicted that is selected by one of the first and second cache eviction algorithms indicated in the output from the at least one machine learning module.