Smart access to a dispersed data storage network

• US 20090094318A1
• Filed: 10/09/2007
• Published: 04/09/2009
• Est. Priority Date: 09/30/2005
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Cleversafe Incorporated (International Business Machines Corporation)
• Inventors: Dhuse, Greg, Volvovski, Ilya, Thornton, Vance, Bellanca, Jamie, Gladwin, S. Christopher
• Granted Patent: US 8,171,101 B2
• Time in Patent Office: Days
• US Class Current: 709/203
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1021 based on client or server locations
• CPC Class Codes: H04L 67/1025 Dynamic adaptation of the criteria on which the server selection is based
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

An improved system for accessing data within a distributed data storage network (“DDSN”) is disclosed. In a system implementing the disclosed invention, traffic is routed to individual slice servers within the DDSN in accordance with objective criteria as well as user-defined policies. In accordance with one aspect of the disclosed invention, when a data segment is written to a DDSN, the segment is divided into multiple data slices, which are simultaneously transmitted to different slice servers. In accordance with another aspect of the disclosed invention, when a data segment is read from a DDSN, a list of slice servers, each containing a data slice that could be used to reconstruct the requested data segment, is assembled, and sorted in accordance with a preference rating assigned to each of the slice servers. Sufficient data slices to reconstruct the data segment are then read in accordance with the preference ranking of the slice servers.

Claims

1. A method of reading data from a dispersed data storage network including a plurality of slice servers, said method operating on at least one computer and comprising the steps of:

i) receiving a read request for a data segment stored on said dispersed data storage network;

ii) assembling a list of said slice servers, wherein each slice server on said list can access at least one data slice associated with said data segment;

iii) assigning a preference rating to each of said slice servers contained on said list; and

iv) retrieving sufficient data slices to assemble said data segment based on said preference rating.

2. The method of claim 1, further comprising the step of sorting said list according to said preference rating.

3. The method of claim 1, wherein said assigning step utilizes at least a slice server'"'"'s most recent response time.

4. The method of claim 1, wherein said assigning step utilizes at least a slice server'"'"'s most recent response time, historical response time, and present load.

5. The method of claim 1, wherein the step of assembling is accomplished in accordance with at least one user-defined policy.

6. A method of writing data to a dispersed data storage network including a plurality of slice servers, said method operating on at least one computer and comprising the steps of:

i) receiving a data segment to be written to said dispersed data storage network;

ii) accessing dispersal parameters including a slice count indicative of a number of data slices said data segment will be divided into, and a redundancy count indicative of the number of data slices that can be lost without comprising the ability to reconstruct said data segment from said dispersed data storage network;

iii) dividing said data segment into a plurality of data slices equal to said slice count;

iv) simultaneously transmitting said plurality of data slices to a plurality of slice servers so that each data slice is transmitted to a separate slice server;

v) storing said plurality of data slices so that each data slice is accessible by a separate slice server; and

vi) monitoring that at least a predetermined number of data slices are successfully stored.

7. The method of claim 6, wherein said monitoring step requires that all of said plurality of data slices are successfully stored.

8. The method of claim 6, wherein said monitoring step requires that at least slice count minus redundancy count data slices are successfully stored.

9. A distributed computer system implementing a dispersed data storage network comprising:

i) a client computer including a first network port coupled to a data network, said client computer originating a read request for a particular data segment;

ii) a plurality of slice servers for storing and retrieving a plurality of data slices, each of said slice servers including a second network port coupled to said data network;

iii) a grid access computer including a third network port coupled to said data network, said grid access computer executing a data assembly algorithm,

wherein the grid access computer (a) receives said read request through said third network port, (b) assembles a list of said slice servers having access to at least one data slice associated with said data segment, (c) assigns a preference rating to each of said slice servers on said list, and (d) retrieves sufficient data slices to assemble said data segment based on said preference rating.

10. The distributed computer system of claim 9, wherein said grid access computer further (e) sorts said list in accordance with said preference rating.

11. The distributed computer system of claim 9, wherein said grid access computer assigns a preference rating to each of said slice servers on said list utilizing at least each of said slice servers'"'"' most recent ping time.

12. The distributed computer system of claim 9, wherein said grid computer assigns a preference rating to each of said slice servers on said list utilizing at least each of said slice servers'"'"' most recent ping time, historical response time, present load, and available storage.

13. The distributed computer system of claim 9, wherein said grid computer assembles a list of said slice servers having access to at least one data slice associated with said data segment in accordance with at least one user-defined policy.

14. A distributed computer system implementing a dispersed data storage network comprising:

i) a client computer including a first network port coupled to a data network, said client computer originating a write request for a particular data segment;

ii) a plurality of slice servers for storing and retrieving a plurality of data slices, each of said slice servers including a second network port coupled to said data network;

iii) a grid access computer with access to a set of dispersal parameters including a slice count indicative of a number of data slices said data segment will be divided into, and a redundancy count indicative of the number of data slices that can be lost without comprising the ability to reconstruct said data segment, said grid access computer including a third network port coupled to said data network and executing a data dispersal algorithm, wherein the grid access computer (a) receives said write request through said third network port, (b) divides said data segment into a plurality of data slices equal to said slice count, (c) simultaneously transmits said plurality of data slices to said plurality of slice servers so that each data slice is transmitted to a separate slice server, and (d) monitors said plurality of slice servers so that at least a predetermined number of data slices are successfully stored.

15. The distributed computer system of claim 14, wherein said grid access computer monitors said plurality of slice servers so that all of said data slices are successfully stored.

16. The distributed computer system of claim 14, wherein said grid access computer monitors said plurality of slice servers so that at least slice count minus redundancy count data slices are successfully stored.