STORAGE ENGINE NODE FOR CLOUD-BASED STORAGE

• US 20130036272A1
• Filed: 08/02/2011
• Published: 02/07/2013
• Est. Priority Date: 08/02/2011
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Nelson, Steven Boyd
• Application Number:
  US13/195,848
• Publication Number:
  US 20130036272A1
• Time in Patent Office: Days
• US Class Current: 711/147
• CPC Class Codes: G06F 16/137 Hash-based content-based indexing of textual data G06F16/31
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• 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

A system includes a storage engine node that includes a processor and a memory coupled to the processor. The memory stores a protocol mapper executable by the processor to convert storage access requests from a local storage protocol to a cloud storage protocol.

Claims

1. A system comprising:

a storage engine node comprising;

a processor; and

a memory coupled to the processor, the memory storing;

a protocol mapper executable by the processor to convert storage access requests from a local storage protocol to a cloud storage protocol; and

a shared memory segment that stores a portion of an index, wherein the shared memory segment is one of a plurality of shared memory segments that collectively store the index.

2. The system of claim 1, wherein each of the plurality of shared memory segments is stored at a distinct storage engine node.

3. The system of claim 1, wherein the memory further stores an indexing system executable by the processor to perform indexing with respect to one or more remote data storage devices based on the index.

4. The system of claim 3, wherein each entry of the index maps a signature of data stored at a particular storage location of the one or more remote data storage devices to a pointer to the particular storage location wherein the pointer corresponds to an address of a shared address space corresponding to the one or more remote data storage devices.

5. The system of claim 4, wherein the one or more remote data storage devices are associated with a cloud storage service.

6. The system of claim 1, wherein the storage access requests include write requests, read requests, or any combination thereof.

7. The system of claim 1, wherein the local storage protocol comprises a fiber channel (FC)-based protocol, a small computer system interface (SCSI)-based protocol, a transport control protocol/internet protocol (TCP/IP)-based protocol, a common internet file system (CIFS)-based protocol, a network file system (NFS)-based protocol, a serial attached SCSI (SAS)-based protocol, or any combination thereof.

8. The system of claim 1, wherein the cloud storage protocol comprises a representational state transfer (REST)-based protocol.

9. The system of claim 1, wherein the memory further stores a second protocol mapper executable by the processor to convert storage access requests received from other storage engine nodes from the cloud storage protocol to the local storage protocol.

10. The system of claim 7, wherein the memory further stores native replication logic configured to convert native storage requests to the cloud storage protocol to supplement local data storage associated with a computing device with cloud-based storage.

11. The system of claim 1, further comprising an access load balancer coupled to the storage engine node and to a second storage engine node, wherein the access load balancer comprises:

an input configured to receive a storage request from a user device, wherein the storage request includes a public token associated with the access load balancer;

a first output coupled to the storage engine node, wherein the first output is associated with a first private token that is assigned to the storage engine node;

a second output coupled to the second storage engine node, wherein the second output is associated with a second private token that is assigned to the second storage engine node; and

load balancing logic executable to;

map the public token to the first private token or to the second private token based on a load balancing method; and

route the access request from the input to the first output or to the second output based on the mapping of the public token.

12. The system of claim 11, wherein the load balancing method comprises a round robin method or a least recently used method.

13. The system of claim 11, wherein the storage engine node and the second storage engine node each execute a common storage engine operating system, wherein the common storage engine operating system comprises a clustered computing operating system.

14. The system of claim 11, wherein the storage engine node and the second storage engine node each execute different storage engine operating systems.

15. The system of claim 11, wherein the storage engine node and the second storage engine node are associated with a first partition index that is accessible to the access load balancer, and wherein a third storage engine node and a fourth storage engine node are associated with a second partition index that is accessible to the access load balancer.

16. A method comprising:

receiving, at a storage engine node of a storage system comprising a plurality of storage engine nodes, a request to write data;

computing a signature of the data to be written;

determining whether the signature is found in an index that is collectively stored in shared memory segments of the plurality of storage engine nodes, wherein each entry of the index maps a signature of data stored at a particular storage location to a pointer to the particular storage location; and

when the signature is not found in the index;

converting the request to write the data from a local storage protocol to a cloud storage protocol;

transmitting the converted request to a cloud-based data storage device; and

adding the signature to the index.

17. The method of claim 16, further comprising, when the signature is found in the index, terminating the request to prevent duplication of storage of the data.

18. The method of claim 16, further comprising:

receiving a second request to read data;

converting the second request from the local storage protocol to the cloud storage protocol; and

transmitting the converted second request to the cloud-based data storage device.

19. A system comprising:

a storage engine node comprising;

a processor; and

a memory coupled to the processor, the memory storing;

a protocol mapper executable by the processor to convert storage access requests from a local storage protocol to a cloud storage protocol; and

native replication logic executable by the processor to convert native storage requests to the cloud storage protocol to supplement local data storage associated with a computing device with cloud-based storage.

20. The system of claim 19, wherein the native replication logic is executable by the processor to receive the native storage requests from the computing device, wherein the local data storage corresponds to a first portion of an address space and wherein the cloud-based storage corresponds to a second portion of the address space.