Emc IP Holding Company LLC

Dell Products Inc.

Michal Yarimi

Itay Keller

Asaf Porath

Dvir Koren

Zeev Shusterman

Yonatan Shtarkman

Neta Peleg

Irit Brener-Shalem

Yonatan Yeshanov

Yuval Miron

Techniques are provided for implementing intelligent load scheduling in a storage system. In an illustrative embodiment, a method comprises collecting a plurality of metrics corresponding to one or more workloads, and processing the plurality of metrics using a machine learning model to predict a load on at least one server. The machine learning model predicts the load on the at least one server based, at least in part, on weights for respective ones of the plurality of metrics generated during training of the machine learning model. In the method, execution by the at least one server of one or more processes in addition to the one or more workloads is scheduled based, at least in part, on the load predicted by the machine learning model.

Intelligent load scheduling in a storage system

Techniques are provided for implementing intelligent defragmentation in a storage system. A storage control system manages a logical address space of a storage volume. The logical address space is partitioned into a plurality of extents, wherein each extent comprises a contiguous block of logical addresses of the logical address space. The storage control system monitors input/output (I/O) operations for logical addresses associated with the extents, and estimates fragmentation levels of the extents based on metadata associated with the monitored I/O operations. The storage control system identifies one or more extents as candidates for defragmentation based at least on the estimated fragmentation levels of the extents.

Intelligent defragmentation in a storage system

An apparatus in an illustrative embodiment comprises at least one processing device comprising a processor coupled to a memory. The apparatus is configured to maintain a snapshot tree data structure having a plurality of volume nodes corresponding to respective ones of (i) a root volume and (ii) multiple snapshots related directly or indirectly to the root volume. The apparatus is further configured to receive a request to read a data item from a given volume offset of a particular one of the volume nodes, to determine a set of data descriptors for the given volume offset, to determine a set of volume nodes of interest for the particular volume node, to determine a contribution set based at least in part on the set of data descriptors and the set of volume nodes of interest, to determine a read address for the data item as a function of the contribution set, and to read the data item from the read address.

Snapshot-enabled storage system implementing algorithm for efficient reading of data from stored snapshots

Techniques are provided for managing snapshot volume trees in a snapshot-enabled data storage system. A snapshot management system is configured to maintain a snapshot volume tree for a storage volume of a storage system, wherein the snapshot volume tree comprises a data structure which comprises a plurality of snapshot volume nodes corresponding to respective ones of (i) a root volume and (ii) multiple snapshots related directly or indirectly to the root volume, wherein the snapshot volume nodes comprise respective generation times which denote times that the respective snapshot volume nodes were created. The generation times are used in conjunction with write times of data items that are written to the snapshot volume nodes to manage snapshot reading and deletion operations.

Management of volume snapshots in a data storage system

Techniques are provided for managing storage device errors during processing of inflight input/output (I/O) operations. A storage control system (e.g., a software-defined storage system) receives an I/O write request which includes data to be written to at least one storage device of a plurality of storage devices managed by the storage control system, and commences a write operation to write the data to the at least one storage device. In response to a storage device I/O error resulting from a failure of the write operation associated with the at least one storage device, the storage control system accesses a logical storage device in a non-volatile system memory device, and writes the data to the logical storage device in the non-volatile system memory device to complete the failed write operation.

Growing community of inventors

Rehovot, Israel

Michal Yarimi