Methods and apparatus for distributing information to multiple nodes
AssigneeEMC Corporation
InventorsStephen J. Todd, Tom Teugels
Life of this patent
Estimating the term…
Abstract
Some embodiments are directed to a technique for storing and/or locating content units stored on an object addressable storage (OAS) system, wherein each content unit is identified by an object identifier. The OAS system may comprise a plurality of zones, each of which stores content units. A mapping process may be defined that maps object identifiers for content units to zones on the OAS system. Thus, the storage location for a content unit on the OAS system may be the zone on the OAS system to which the object identifier for the content unit maps.
Claims20
A method of processing content in a computer system comprising at least one host and at least one storage system that comprises a plurality of zones, wherein the at least one host accesses content units stored on the at least one storage system using object identifiers, wherein the computer system maps each object identifier to one of the plurality of zones on the at least one storage system to which the corresponding content unit is allocated for storage, wherein the at least one storage system comprises a plurality of nodes, each of which is configured to receive and process access requests, wherein each of the plurality of nodes stores first information identifying which object identifiers map to which zones on the at least one storage system, and wherein the method comprises acts of:
- sending to each of the plurality of nodes second information, to replace the first information, that identifies which object identifiers map to which zones on the at least one storage system;
- sending to each of the plurality of nodes an indication of a future time associated with the second information;
- receiving, from each of the plurality of nodes, an acknowledgement of receipt of the second information;
- and
- in response to receiving the acknowledgement from each of the plurality of nodes, sending to each of the plurality of nodes a command to stop using the first information to map object identifiers to zones and to begin using the second information to map object identifiers to zones at the future time.
At least one computer readable medium encoded with instructions that, when executed in a computer system comprising at least one host and at least one storage system that comprises a plurality of zones, perform a method of processing content wherein the at least one host accesses content units stored on the at least one storage system using object identifiers, wherein the computer system maps each object identifier to one of the plurality of zones on the at least one storage system to which the corresponding content unit is allocated for storage, wherein the at least one storage system comprises a plurality of nodes, each of which is configured to receive and process access requests, wherein each of the plurality of nodes stores first information identifying which object identifiers map to which zones on the at least one storage system, and wherein the method comprises acts of:
- sending to each of the plurality of nodes second information, to replace the first information, that identifies which object identifiers map to which zones on the at least one storage system;
- sending to each of the plurality of nodes an indication of a future time associated with the second information;
- receiving, from each of the plurality of nodes, an acknowledgement of receipt of the second information;
- and
- in response to receiving the acknowledgement from each of the plurality of nodes, sending to each of the plurality of nodes a command to stop using the first information to map object identifiers to zones and to begin using the second information to map object identifiers to zones at the future time.
At least one storage system configured to process access requests for content units from at least one host computer which identify content units using object identifiers, the at least one storage system comprising:
- a plurality of zones, each of which is configured to store content units;
- a plurality of nodes, each of which is configured to receive and process access requests, wherein each of the plurality of nodes stores first information identifying which object identifiers map to which zones on the at least one storage system;
- and
- at least one controller that:
- sends to each of the plurality of nodes second information, to replace the first information, that identifies which object identifiers map to which zones on the at least one storage system;
- sends to each of the plurality of nodes an indication of a future time associated with the second information;
- receives, from each of the plurality of nodes, an acknowledgement of receipt of the second information;
- and
- in response to receiving the acknowledgement from each of the plurality of nodes, sends to each of the plurality of nodes a command to stop using the first information to map object identifiers to zones and to begin using the second information to map object identifiers to zones at the future time.
Dependencies are inferred from the claim text.