Immediate copy target pull of volume data
AssigneeInternational Business Machines Corporation
InventorsFrederick G. Stock, Joseph M. Swingler
Life of this patent
Estimating the term…
Abstract
A composite storage server employs a plurality of distributed storage sites including a mount site, a source copy site and one or more copy target sites for executing a storage of volume data within the composite storage server. In operation, an immediate copy target site receives a communication of a source copy complete message from the mount site, the source copy complete message being indicative of the source copy site having a complete copy of the volume data. Based on the source copy complete message, the immediate copy target site pulls the volume data from a peer distributed storage site having a complete copy of the volume data (e.g., the source copy site, the mount site or another copy target site).
Claims19
A composite storage server, comprising:
- a plurality of interconnected peer distributed storage sites including a mount site, a source copy site and at least one copy target site for executing a host job involving a storage of volume data within the composite storage server;
- wherein the mount site is operable to communicate a source copy complete message to an immediate copy target site, the source copy complete message being indicative of the source copy site having a complete copy of the volume data, wherein the source copy site is designated by the mount site for storing a primary copy of the volume data;
- wherein the immediate copy target site queries the peer distributed storage sites to determine one of the peer distributed storage sites having the complete copy of the volume data to pull in response to the source copy complete message, wherein the immediate copy target site pulls the volume data from the determined peer distributed storage site having the complete copy of the volume data, and wherein the immediate copy target site pulls and stores the volume data prior to a complete execution of the host job:
- wherein the immediate copy target site is further operable to communicate an immediate copy complete message to the mount site, the immediate copy complete message being indicative of the immediate copy target site having the complete copy of the volume data;
- and
- wherein the mount site is further operable to indicate a redundancy status of the composite storage server as having an adequate redundancy based on the mount site receiving the immediate copy complete message from the immediate copy target site prior to an expiration of a redundancy copy timer.
In a composite storage server comprising a plurality of interconnected peer distributed storage sites including a mount site, a source copy site and at least one copy target site for executing a host job involving a storage of volume data within the composite storage server, an immediate copy target site comprising:
- a processor;
- and
- a memory storing instructions operable with the processor, the instructions executed for:
- receiving a communication of a source copy complete message from the mount site, the source copy complete message being indicative of the source copy site having a complete copy of the volume data, wherein the source copy site is designated by the mount site for storing a primary copy of the volume data;
- querying by the immediate copy target site the peer distributed storage sites to determine one of the peer distributed storage sites having the complete copy of the volume data to pull in response to the source copy complete message;
- pulling, by the immediate copy target site, the volume data from the determined peer distributed storage site having the complete copy of the volume data, wherein the immediate copy target site pulls and stores the volume data prior to a complete execution of the host job;
- communicating, by the immediate copy target site, an immediate copy complete message to the mount site, the immediate copy complete message being indicative of the immediate copy target site having the complete copy of the volume data;
- and
- indicating, by the mount site, that a redundancy status of the composite storage server has an adequate redundancy based on the mount site receiving the immediate copy complete message from the immediate copy target site prior to an expiration of a redundancy copy timer.
In a composite storage server comprising a plurality of interconnected peer distributed storage sites including a mount site, a source copy site and at least one copy target site for executing a host job involving a storage of volume data within the composite storage server, a method of operating an immediate copy target site comprising:
- the immediate copy target site receiving a communication of a source copy complete message from the mount site, the source copy complete message being indicative of the source copy site having a complete copy of the volume data, wherein the source copy site is designated by the mount site for storing a primary copy of the volume data;
- the immediate copy target site querying the peer distributed storage sites to determine one of the peer distributed storage sites having the complete copy of the volume data to pull in response to the source copy complete message;
- the immediate copy target site pulling the volume data from the determined peer distributed storage site having the complete copy of the volume data, and wherein the immediate copy target site pulls and stores the volume data prior to a complete execution of the host job;
- the immediate copy target site communicating an immediate copy complete message to the mount site, the immediate copy complete message being indicative of the immediate copy target site having the complete copy of the volume data;
- and
- the mount site indicating a redundancy status of the composite storage server as having an adequate redundancy based on the mount site receiving the immediate copy complete message from the immediate copy target site prior to an expiration of a redundancy copy timer.
A composite storage server, comprising:
- a plurality of interconnected peer distributed storage sites including a mount site, a source copy site and at least one copy target site for executing a host job involving a storage of volume data within the composite storage server;
- wherein the mount site is operable to communicate a source copy complete message to an immediate copy target site, the source copy complete message being indicative of the source copy site having a complete copy of the volume data, wherein the source copy site is designated by the mount site for storing a primary copy of the volume data;
- wherein the immediate copy target site queries the peer distributed storage sites to determine one of the peer distributed storage sites having the complete copy of the volume data to pull in response to the source copy complete message, and wherein the immediate copy target site pulls the volume data from the determined peer distributed storage site having the complete copy of the volume data, and wherein the immediate copy target site pulls and stores the volume data prior to a complete execution of the host job;
- and
- wherein a host job manager performs:
- in response to determining that the host job is executing in one of the distributed storage sites comprising the source copy site, transition to an idle state to wait for activation by a host to function as the source copy site for storing a primary copy of the volume;
- in response to determining that the host job is executing in one of the distributed storage sites comprising the immediate copy target site, transition to an idle state to wait for activation to function as an immediate copy target site that stores a copy from the source copy site prior to a complete execution of the host job;
- and
- in response to determining that the host job is executing in one of the distributed storage sites comprising a deferred copy target site, transition to an idle state to wait for activation to function as an immediate copy target site that stores a copy from the source copy site subsequent to a complete execution of the host job.
Dependencies are inferred from the claim text.