International Business Machines Corporation

Strato Scale Ltd.

Aharon Lazar

Shay Haim Akirav

Gil E Paz

Itay Maoz

Elena Drobchenko

Yariv Bachar

Evgeny Budilovsky

Ofer Leneman

Tzafrir Z Taub

Dov N Hepner

Aviv Caro

Dan Melamed

Samuel Krikler

Asaf Ekshtein

Shiri Alexander

Neville Yates

Ofer Lenaman

Yael Feldmann

A method includes storing data generated in a source node by sending write requests to multiple destination nodes. The destination nodes are requested to create snapshots of the data. The write requests are marked at the source node with marks that indicate to each destination node which of the write requests are pre-snapshot write requests that were issued before a snapshot request for a snapshot that the destination node is currently storing, and which of the write requests are post-snapshot write requests that were issued after the snapshot request for the snapshot that the destination node is currently storing. The snapshots are synchronized with one another at the destination nodes based on the marks.

Synchronization of snapshots in a distributed storage system

Subsequent to a storage operation performed on the source instance by a source component, a synchronization message is sent to a replicated component for the replicated instance. The synchronization message is stored locally in a persistent storage location associated with the source component along with an indicator representative of a time the storage operation was performed. Pursuant to receipt of the synchronization message by the replicated component, the replicated component is updated to a dirty state to indicate a lack of full synchronization between the source and replicated instances.

Synchronization of sequential access storage components with backup catalog

Pursuant to execution of a predetermined failback policy, if a storage device is not preexistent in a source storage system operable in the computing environment, and an owner of the storage device is one of a DR storage system and a storage system having previously withdrawn from a replication grid, and the DR storage system has authorization to transfer ownership of the storage device by a replacement operation, and an instance of the storage device is replicated to the source storage system to generate a replicated instance of the storage device on the source storage system.

Disaster recovery failback

Visibility of a data storage entity on a first storage system is switched to a replicated data storage entity on a second storage system. Data from the data storage entity is replicated from the first storage system to the second storage system using a common serial number. The data storage entity is hidden from the first storage system by concealing the common serial number. An ejection of the data storage entity from the first storage system is automated, and the replicated data storage entity is introduced to the second storage system.

Switching visibility between virtual data storage entities

Various embodiments for synchronization of source and replicated instances of sequential access storage components in a computing environment by a processor device are provided. A replication synchronization table (RST) of a source instance is updated at a source component with the compacted information. Based on the RST of the source instance and the RST of the replicated instance, data representative of a most recent position on the source component from which data should be transmitted to a replicated component to achieve a full synchronization is determined.

Synchronization of replicated sequential access storage components

A replication synchronization table (RST) is configured for the source instance. The RST is adapted for recording compacted information for at least one of a truncate and an erase operation applied on the source instance. The RST of the source instance is updated at a source component with the compacted information. The updated RST is transmitted from the source instance to the replicated instance. Based on the RST of the source instance and the RST of the replicated instance, data representative of a most recent position on the source component from which data should be transmitted to a replicated component to achieve a full synchronization is determined.

Pursuant to execution of a predetermined failback policy, if a storage device is not preexistent in a source storage system operable in the computing environment, and an owner of the storage device is one of a DR storage system and a storage system having previously withdrawn from a replication grid, and the DR storage system has authorization to transfer ownership of the storage device by a replacement operation, an instance of the storage device is replicated to the source storage system to generate a replicated instance of the storage device on the source storage system, and ownership of the storage device is designated as the source storage system for each of the instance and the replicated instance of the storage device.

Subsequent to a storage operation performed on the source instance by a source component, a synchronization message is sent to a replicated component for the replicated instance. The synchronization message is stored locally in a persistent storage location associated with the source component along with an indicator representative of a time the storage operation was performed. Pursuant to receipt of the synchronization message by the replicated component, the replicated component is updated to a dirty state to indicate a lack of full synchronization between the source and replicated instances. Receipt of the synchronization message is acknowledged by the replicated component by sending a dirty state acknowledgement. If, during a subsequent recovery process, the dirty state is not yet cleared, the replicated component is deemed ineligible for participation in a full restoration operation.

Growing community of inventors

Kiryat Ono, Israel

Aharon Lazar