Abstract: One or more aspects of the present disclosure relate to dynamic compression engine management. In embodiments, statistics corresponding to an input/output (IO) workload received by a storage array are collected. Additionally, statistics corresponding to one or more compression cards of the storage array are collected. Further, one or more compression engines within the one or more compression cards of the storage array are dynamically activated or deactivated based on the IO workload and compression hardware statistics.

Abstract: One or more aspects of the present disclosure relate to maximizing data migration bandwidth. In embodiments, one or more network characteristics corresponding to network communications between a first storage array and a second storage array is determined. Further, one or more network metrics of at least one acknowledgment communication from the second storage array to the first storage array can be analyzed. Additionally, one or more input/output (IO) messages are transmitted from the first storage array to the local storage array during an acknowledgment period corresponding to receipt of the at least one acknowledgment communication from the second storage array by the first storage array based on the one or more network metrics.

Abstract: Metadata pages used exclusively for maintenance IOs associated with snapshots, data migration, or data replication are paged-out according to a time series model forecasted TTL. Metadata pages used for production IOs are paged-out using an LRU algorithm. The forecasted TTL corresponds to the expected time during which the metadata pages will be needed to service the maintenance IOs.

Abstract: Methods and apparatuses for improving storage array performance by using time-series characteristics on both a local storage area (extent-level, or track-level) basis as well as global, system level are disclosed. The time series characteristics are input to an aging model where optimal aging times, on a per-extent basis, are forecasted according to the local and global characteristics. The aging model may be a multivariate time-series model. Incoming write requests may be destaged from cache memory according to the forecasted optimal aging times for the storage extents.

Abstract: A method for use in a storage system, comprising: detecting that a backend track is required to be deleted, the backend track currently being in an allocated state; identifying a repetition probability score for data that is stored in the backend track; detecting whether the repetition probability score satisfies a predetermined condition; when the predetermined condition is satisfied, transitioning the backend track from an allocated state into a delayed release state, removing any associations between the backend track and one or more frontend tracks, waiting for a predetermined waiting period, and transitioning the backend track from the delayed release state into a free state, wherein the backend track is transitioned into the free state only if no write requests are received at the storage system which include user data matching the data that is currently stored in the backend track.

Abstract: Metadata pages used exclusively for maintenance IOs associated with snapshots, data migration, or data replication are paged-out according to a time series model forecasted TTL. Metadata pages used for production IOs are paged-out using an LRU algorithm. The forecasted TTL corresponds to the expected time during which the metadata pages will be needed to service the maintenance IOs.

Abstract: A method comprising: generating a write-on-write distribution for a plurality of data units in a data entity, the write-on-write distribution including a plurality of buckets, each bucket corresponding to a different write-on-write time range, each bucket having a respective bucket size that is based on a count of ones of the plurality of data units whose respective predicted write-on-write times fall within the bucket's write-on-write time range; identifying one or more utilization metrics of a storage system where the data entity is hosted; classifying a representative bucket size of the write-on-write distribution and the one or more utilization metrics with a machine learning model to predict a replication period for the data entity, the representative bucket size being a bucket size within a predetermined distance from a mean bucket size of the write-on-write distribution; and configuring the storage system to replicate the data entity in accordance with the replication period.

Abstract: Methods and apparatuses for improving storage array performance by using time-series characteristics on both a local storage area (extent-level, or track-level) basis as well as global, system level are disclosed. The time series characteristics are input to an aging model where optimal aging times, on a per-extent basis, are forecasted according to the local and global characteristics. The aging model may be a multivariate time-series model. Incoming write requests may be destaged from cache memory according to the forecasted optimal aging times for the storage extents.

Abstract: A method for creating snapshots is disclosed. The method includes collecting a first resource usage metric of a first storage system and a second resource usage metric for one or more second storage systems, detecting a command to create a snapshot of a storage entity, and detecting when the snapshot would be a base snapshot. When the snapshot is going to be a base snapshot, a selection policy is evaluated to identify the location where the snapshot is going to be created after which the snapshot is created at the identified location. When the snapshot would not be a base snapshot, the snapshot is stored at a location where snapshots of the storage entity are already being created.

Abstract: For a source device with an activated snapshot, time series-based prediction of source device write IO operations is implemented, on a per-extent basis, to predict a subset of the source device extents that are likely to be hot (receive write IO operations) during an upcoming time window. Snapshot tracks corresponding to tracks of the predicted hot extents are pre-deduplicated, to accelerate write IO operations on the source device. In instances where the time series-based prediction correctly predicts write IO operations on tracks of an extent, and the tracks of the extent of the source device are pre-deduplicated on the snapshot, it is possible to implement the write IO operations as a redirect on write operation, without first replicating the original track of source data for use by the snapshot. Write IO operations on tracks that are not pre-deduplicated are implemented as copy on write operations.

Abstract: Metadata page prefetch processing for incoming IO operations is provided to increase storage system performance by reducing the frequency of metadata page miss events during IO processing. When an IO is received at a storage system, the IO is placed in an IO queue to be scheduled for processing by an IO processing thread. A metadata page prefetch thread reads the LBA address of the IO and determines whether all of the metadata page(s) that will be needed by the IO processing thread are contained in IO thread metadata resources. In response to a determination that one or more of the required metadata pages are not contained in IO thread metadata resources, the metadata page prefetch thread instructs a MDP thread to move the required metadata page(s) from metadata storage to IO thread metadata resources. The IO processing thread then implements the IO operation using the prefetched metadata.

Abstract: Metadata page prefetch processing for incoming IO operations is provided to increase storage system performance by reducing the frequency of metadata page miss events during IO processing. When an IO is received at a storage system, the IO is placed in an IO queue to be scheduled for processing by an IO processing thread. A metadata page prefetch thread reads the LBA address of the IO and determines whether all of the metadata page(s) that will be needed by the IO processing thread are contained in IO thread metadata resources. In response to a determination that one or more of the required metadata pages are not contained in IO thread metadata resources, the metadata page prefetch thread instructs a MDP thread to move the required metadata page(s) from metadata storage to IO thread metadata resources. The IO processing thread then implements the IO operation using the prefetched metadata.

Abstract: In asynchronous remote replication, write IOs are accumulated in capture cycles and sent to a remote storage system in transmit cycles. In order to cause metadata cache hits at the remote storage system, write IO data and associated metadata hints such as logical block addresses being updated are sent in successive cycles. The metadata hints, which are received at the remote storage system before the corresponding write IO data, are used to prefetch metadata associated with the logical block addresses being updated to replicate the write IO.

Abstract: Selective packing of small block write operations is implemented prior to compression, to improve compression efficiency and hence reduce bandwidth requirements of a Remote Data Replication (RDR) facility. Compression characteristics of write IO operations are forecast, and write IO operations with similar forecast compression characteristics are pooled. Write IO operations are also grouped according to extent, device, and storage group. Write operations from a given compression pool are then preferentially selected from the extent-level grouping, next from the device-level grouping, and then from the SG-level grouping, to create an IO package. The IO package is then compressed and transmitted on the RDR facility. By creating an IO package prior to compression, it is possible to achieve greater compression than would be possible if each individual write IO operation were to be individually compressed to thereby reduce network bandwidth of the RDR facility.

Abstract: One or more aspects of the present disclosure relate to performant destaging of write pending (WP) data to disk. In embodiments, an input/output (IO) workload is received at a storage array. Additionally, the IO workload can include an IO request with a random write request. Further, a write destage context for write-pending (WP) data corresponding to the random write request can be generated by a data services engine of the storage array. In addition, using the write destage context, a disk adapter (DA), at a backend of the storage array, is enabled to destage write-pending (WP) data without reading from a target write location of the random write request on a storage device on the storage array.

Abstract: One or more aspects of the present disclosure relate to maximizing data migration bandwidth. In embodiments, one or more network characteristics corresponding to network communications between a first storage array and a second storage array is determined. Further, one or more network metrics of at least one acknowledgment communication from the second storage array to the first storage array can be analyzed. Additionally, one or more input/output (IO) messages are transmitted from the first storage array to the local storage array during an acknowledgment period corresponding to receipt of the at least one acknowledgment communication from the second storage array by the first storage array based on the one or more network metrics.

Abstract: A remote data replication facility includes a primary storage array and a backup storage array, on which tracks of data are replicated from the primary storage array to the backup storage array as they are received by the primary storage array. Remote data verification is implemented on the remote data replication facility by comparing track fingerprints, track temporal write metadata, and track spatial write metadata, for a given track on the primary storage array, with corresponding track fingerprints, track temporal write metadata, and track spatial write metadata, for the given track on the backup storage array. If any difference is determined in the combination of track fingerprints, track temporal write metadata, track spatial write metadata, for a given track, the integrity of the data at the backup storage array is not verified for the track.

Abstract: One or more aspects of the present disclosure relate to performant destaging of write pending (WP) data to disk. In embodiments, an input/output (IO) workload is received at a storage array. Additionally, the IO workload can include an IO request with a random write request. Further, a write destage context for write-pending (WP) data corresponding to the random write request can be generated by a data services engine of the storage array. In addition, using the write destage context, a disk adapter (DA), at a backend of the storage array, is enabled to destage write-pending (WP) data without reading from a target write location of the random write request on a storage device on the storage array.

Abstract: The present disclosure relates to applying a data deduplication layer on top of an asynchronous remote replication protocol. In embodiments, a remote replication group including one or more of a storage array's logical unit numbers (LUNs) can be established. Further, at a remote system, —the remote replication group can be remotely replicated on a per LUN basis using a deduplication fingerprint of each LUN's data tracks corresponding to the remote replication group.

Abstract: The present disclosure relates to applying a data deduplication layer on top of an asynchronous remote replication protocol. In embodiments, a remote replication group including one or more of a storage array's logical unit numbers (LUNs) can be established. Further, at a remote system, —the remote replication group can be remotely replicated on a per LUN basis using a deduplication fingerprint of each LUN's data tracks corresponding to the remote replication group.