International Business Machines Corporation

Globalfoundries Inc.

Ofer Peretz

Asaf Yeger

Eran Raichstein

Gregory John Tevis

Christopher Zaremba

David Gregory Van Hise

Michael Sternberg

Uri Wolloch

Eran Tzabari

Oren Wolf

Ovadi Somech

Boris Litvinov

Yaron Kraus

For dynamic capacity allocation of stripes in cluster-based storage system, a storage controller identifies a storage stripe in a cluster-based storage system that has a smallest stripe score of a plurality of storage stripes. The storage controller also identifies a storage stripe in the cluster-based storage system that has a largest stripe score of the plurality of storage stripes, and reallocates, in response to a difference between the largest stripe score and the smallest stripe score being less than a threshold value, free-space capacity from a donator stripe to a receiver stripe.

Dynamic capacity allocation of stripes in cluster based storage systems

Machines, systems and methods for performing intermediate data backups, the method comprising monitoring data updates to one or more data blocks in at least a target data storage medium, wherein the target data storage medium is subject to an incremental data backup routine at prescheduled time intervals; in response to determining that said at least one or more data blocks is updated prior to a prescheduled time interval for the incremental data backup routine, performing one or more intermediate data backups to store data from the updated data blocks to at least one backup data storage medium; and in response to determining that said at least one or more data blocks is updated prior to the prescheduled time interval for the incremental data backup routine but after the last of the intermediate data backups, copying data on one or more updated data blocks after the last of the intermediate data backups to the backup data storage medium.

Intermediate data backup in an incremental snapshot data backup system to reduce data backup time

Coordinating point-in-time snapshots among multiple computing platforms by receiving a notification from a first computing platform agent indicating a first computing platform snapshot time, receiving a notification from a second computing platform agent indicating a second computing platform snapshot time, determining that second computing platform snapshot time is later than the first computing platform snapshot time, notifying the first computing platform agent of the second computing platform snapshot time, and receiving from the first computing platform agent a report of any location in the first computing platform's data storage to which data were written after the first computing platform snapshot time and responsive to a write request that was made prior to or including the second computing platform snapshot time.

Coordinated point-in-time snapshots of multiple computing platforms

Machines, systems and methods for backing up data, the method comprising retrieving identifying information about target data storage volumes from which target data is to be backed up; communicating the identifying information to a proxy application installed on a backup computing system utilized to backup the target data stored on the target data storage volumes, wherein the proxy application uses the identifying information to place the target data storage volumes in a first state in which write operations may not be performed to the target data storage volumes, wherein the proxy application causes snapshots to be taken of the target data storage volumes of a first type and a second type, in response to determining that the target data storage volumes are placed in the first state.

Data backup across physical and virtualized storage volumes

Described is a method, system, and computer program product for selecting a data repository within a computing environment. The data repository can exist on multiple target computing systems as a logical entity called a volume or a file system. When selecting at least one data repository, a data protection system of the computing environment analyzes the incoming data from a source computing system, in conjunction with information describing the data repositories. The data repository information, or repository characteristic information, can be stored on a repository volume table (RVT) of the data protection system. By determining a weighted selection score from the analysis of all the information available to the data protection system, the data protection system can intelligently select a data repository for storing data from the source computing system. Further, by tracking repository characteristic information on a continuous or periodic basis, the data protection system can enable emulated views of data that has been stored on the data repositories of the computing environment.

Data repository selection within a storage environment

Growing community of inventors

Haifa, Israel

Ofer Peretz