Reading and writing during cluster growth phase

• US 8,447,833 B2
• Filed: 03/06/2012
• Issued: 05/21/2013
• Est. Priority Date: 04/19/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Nightingale, Edmund B., Elson, Jeremy E., Howell, Jonathan R.
• Primary Examiner(s): DINH, KHANH Q
• Application Number:
  US13/412,944
• Publication Number:
  US 20120166590A1
• Time in Patent Office: 441 Days
• Field of Search: 709/203, 709/227, 709/228, 709/217, 709/220, 709/229
• US Class Current: 709/217
• CPC Class Codes: G06F 11/10 Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 2211/104 Metadata, i.e. metadata associated with RAID systems with parity
• CPC Class Codes: G06F 3/06 Digital input from, or digital output to, record carriers , e.g. RAID, emulated record carriers or networked record carriers
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

A client device configured to write to both a growth server and a live server replica that the growth server is replacing during a growth phase is described herein. The client device first determines the growth server designated to replace a corresponding live server replica of a set of server replicas associated with a storage index. The client device then transmits a write request associated with the storage index to the set of server replicas and to the growth server. The client device may perform the determining based on storage assignment mappings. The storage assignment mappings are provided to the client device by a metadata server while the growth server retrieves data associated with the storage index from the live server replica.

Claims

1. A method comprising:

transmitting, by a client device, an initial write request to a set of servers replicas;

receiving, by the client device, in response to the initial write request an indication that a version of storage assignment mappings utilized by the client device is not a most recent version;

retrieving the most recent version of the storage assignment mappings;

determining, by the client device, one or more growth servers designated to replace a corresponding one or more server replicas of the set of server replicas associated with a storage index; and

transmitting, by the client device, a write request associated with the storage index to the set of server replicas and the one or more growth servers.

2. The method of claim 1, further comprising reading data associated with the storage index from the set of server replicas without reading data from the growth servers.

3. The method of claim 1, further comprising performing the determining based on storage assignment mappings associating each of a plurality of storage indices with a set of server replicas.

4. The method of claim 3, wherein the storage assignment mappings are a table with:

rows representing storage indices and columns representing server replicas and growth servers, and the growth servers are represented in one or both of first or second ones of the columns, or

columns representing storage indices and rows representing server replicas and growth servers, and the growth servers are represented in one or both of first or second ones of the rows.

5. The method of claim 3, wherein the storage assignment mappings are a table with:

rows representing storage indices and columns representing server replicas and growth servers, and one or more of the columns are associated with a growth indicator that indicates that those one or more columns include identifiers of growth servers in their associated fields, or

columns representing storage indices and rows representing server replicas and growth servers, and one or more of the rows are associated with a growth indicator that indicates that those one or more rows include identifiers of growth servers in their associated fields.

6. The method of claim 1, further comprising determining that a growth indicator associated with the storage index indicates which growth servers are designated to replace given server replicas associated with the storage index, and, in response, performing the determining of the growth servers.

7. The method of claim 1, wherein the determining includes examining tags associated with identifiers of the one or more server replicas, the tags including identifiers of the corresponding one or more growth servers.

8. The method of claim 1, further comprising determining the one or more server replicas designated to be replaced by the one or more growth servers by examining tags associated with identifiers of the one or more growth servers, the tags including identifiers of the designated one or more server replicas.

9. The method of claim 1, wherein the write request is associated with a byte sequence that utilizes a subset of server replicas associated with the storage index and the subset of server replicas comprises the set of server replicas.

10. The method of claim 9, wherein the storage index is associated with a plurality of growth servers, and the method comprises determining a subset of the plurality of growth servers as the one or more growth servers designated to replace the one or more server replicas.

11. A client device comprising:

a processor; and

a plurality of instructions configured to be executed by the processor to perform operations including;

transmitting an initial write request to a set of servers replicas;

receiving in response to the initial write request an indication that a version of storage assignment mappings utilized by the client device is not a most recent version; and

retrieving the most recent version of the storage assignment mappings

determining whether one or more growth servers are designated to replace a corresponding one or more server replicas of set of server replicas associated with a storage index; and

in response to determining that growth servers are designated to replace corresponding server replicas, writing data associated with the storage index to the set of server replicas and the one or more growth servers, and reading data associated with the storage index from the set of server replicas without reading data from the growth servers.

12. The client device of claim 11, wherein the operations further include, upon replacement of the one or more server replicas with the corresponding growth servers with respect to the storage index, receiving an updated version of storage assignment mappings identifying the one or more growth servers as server replicas included in the set of server replicas associated with the storage index.

13. The client device of claim 12, wherein the operations further include, after receiving the updated version of the storage assignment mappings, transmitting write requests associated with the storage index to the set of server replicas without transmitting the write requests to the replaced one or more server replicas.

14. One or more computer-readable storage media including a plurality of computer-executable instructions stored thereon and configured to program a metadata server to perform operations comprising:

instructing a growth server designated to replace a live server with respect to a storage index to retrieve data stored on the live server both from the live server and from other server replicas associated with the storage index;

while the growth server retrieves the data, providing storage assignment mappings associating the live server, the growth server, and other server replicas with the storage index to one or more client devices to enable the client devices to transmit first write requests associated with the storage index to the live server, the growth server, and the other server replicas;

receiving an indication from the growth server that the growth server has retrieved the data; and

in response to receiving the indication, providing to the one or more client devices updated storage assignment mappings replacing the live server with the growth server with respect to the storage index, the updated storage assignment mappings enabling the client devices to transmit second write requests to the growth server and the other server replicas without transmitting the second write requests to the live server.

15. The one or more computer-readable storage media of claim 14, wherein the updated storage assignment mappings comprise updated storage assignment mappings for a subset of storage indices included in a larger set of storage assignment mappings, each storage index having an associated version identifier and a number of associated server replicas.

16. The one or more computer-readable storage media of claim 14, wherein the operations further comprise generating the storage assignment mappings, the generating including associating the growth server with the storage index and setting a version identifier associated with the storage assignment mappings.

17. The one or more computer-readable storage media of claim 14, wherein the operations further comprise, responsive to receiving the indication, instructing the live server to release storage resources associated with the storage index.

18. The one or more computer-readable storage media of claim 14, wherein the growth server is also designated to replace a different live server with respect to a different storage index.

19. The one or more computer-readable storage media of claim 14, wherein the operations further comprise initiating a failure recovery process in response to failure of the metadata server, the failure recovery process including:

retrieving first identifications from live servers in a system of first storage indices associated with the live servers and second identifications from growth servers in a system of second storage indices associated with the growth servers, the second identifications from the growth servers being accompanied by indications that the growth servers are in growth states with respect to the second storage indices; and

generating storage assignment mappings based on the retrieved first identifications, second identifications, and indications.