ACQUISITION AND WRITE VALIDATION OF DATA OF A NETWORKED HOST NODE TO PERFORM SECONDARY STORAGE

• US 20100169282A1
• Filed: 12/26/2008
• Published: 07/01/2010
• Est. Priority Date: 06/01/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: InMage Systems, Inc. (Microsoft Corporation)
• Inventors: Atluri, Rajeev, Inala, Sam, Tirumala, Anup S.
• Granted Patent: US 8,224,786 B2
• Time in Patent Office: Days
• US Class Current: 707/655
• CPC Class Codes: G06F 16/2308 Concurrency control transaction processing G06F9/466

Abstract

Methods and a system to acquire and perform write validation of data generated by one or more networked host nodes to perform secondary storage are disclosed. According to one embodiment, a method to passively acquire and perform write validation of data generated by one or more networked host nodes to perform secondary storage in a SAN-based data storage and recovery network includes generating data to store in primary storage. The method further includes generating metadata describing the data generated to store in primary storage, sending the data and metadata to a primary SAN storage device, acquiring passive access to data traveling a data path between a generating node and the primary SAN storage device, the data mirrored over an access line to a secondary storage server. The method further includes receiving, at the secondary storage server, an exact copy of a data stream that passes a splitter.

Claims

1. A method to passively acquire and perform write validation of data generated by a networked host node to perform secondary storage in a SAN-based data storage and recovery network, comprising:

(a) generating data to store in primary storage;

(b) generating metadata describing the data generated to store in primary storage;

(c) sending the data and metadata to a primary SAN storage device;

(d) acquiring passive access to data traveling a data path between a generating node and the primary SAN storage device the data mirrored over an access line to a secondary storage server;

(e) receiving, at the secondary storage server, an exact copy of a data stream that passes a splitter; and

(f) comparing metadata associated with the data against additional metadata to validate or invalidate a write data, wherein the additional metadata is stored in at least one of a local location and an external location.

2. The method of claim 1 wherein in step (a) the data is generated by a LAN connected PC or a dedicated server node.

3. The method of claim 1 wherein in step (a) a primary storage medium is comprised of a RAID unit accessible through a network switch, and wherein the network switch is comprised of at least one of a Fibre Channel switch and an Ethernet hub.

4. The method of claim 1 wherein in step (b) the metadata describes at least a generating node ID, a destination ID of the primary SAN storage device, a write offset location in primary storage, a length of a payload, and checksum data.

5. The method of claim 1 wherein in step (c) the data is sent as a series of data frames conforming to at least one of a SCSI protocol and an Ethernet protocol.

6. The method of claim 1 wherein in step (d) data path splitting is achieved using a hardware data splitter of at least one of an optical type and an electrical type, depending on the type of network line used.

7. The method of claim 1 wherein in step (e) the secondary storage server is at least one of a dedicated server node and a PC node, wherein the data is passively received by at least one of a specially adapted line card installed in the secondary storage server and a network adaptor card.

8. The method of claim 1 wherein in step (f) metadata comparison is performed on a line card adapted to receive the data.

9. The method of claim 1 wherein in step (f) metadata comparison is performed in a cache system.

10. The method of claim 1 wherein the SAN-based data storage and recovery network is provided as a remote service accessible over a network.

11. The method of claim 1 wherein in step (f) metadata comparison is performed at a host prior to completion of a write operation to prevent a redundant save.

12. The method of claim 1 wherein in step (d) data path splitting is performed at the one or more host nodes.

13. The method of claim 1 wherein metadata is created in a host and is received by the secondary storage server after an associated data frame

14. The method of claim 1 wherein metadata is sent to a host server via LAN.

15. The method of claim 1 wherein server memory is used by an external device to store metadata about data frames written to secondary storage to be compared with local metadata.

16. The method of claim 1 wherein data stored in secondary storage is converted using at least one of a sparse file utility and a compression algorithm.

17. A system to passively acquire and perform write validation of data generated by a networked host node to store in secondary storage in a SAN-based data storage and recovery network, comprising:

(a) one or more networked host nodes to generate data to store in primary storage;

(b) a metadata generator module to generate metadata describing the data generated to store in primary storage, wherein the metadata generator is comprised of at least one of the one or more networked host nodes, a host node client, a data splitter, a line card, and a secondary storage server;

(c) a primary SAN storage device to receive the data and metadata;

(d) a splitter to acquire passive access to data traveling a data path between a generating node and the primary SAN storage device and to mirror the data over an access line to a secondary storage server;

(e) the secondary storage server to receive an exact copy of a data stream that passes the splitter; and

(f) a metadata comparison module to compare metadata associated with an actual data against additional metadata to validate or invalidate write data, wherein the additional metadata is stored in at least one of a local location and an external location, and wherein the metadata comparison module is comprised of at least one of the one or more networked host nodes, a host node client, a data splitter, a line card, and a secondary storage server.

18. The system of claim 17 wherein the one or more networked host nodes are comprised of at least one of a LAN connected PC and a dedicated server node.

19. The system of claim 17 wherein a primary storage medium is comprised of a RAID unit accessible through a network switch, and wherein the network switch is comprised of at least one of a Fibre Channel switch and an Ethernet hub.

20. The method, comprising:

(a) forming a networked host node to generate data to store in primary storage in a system to passively acquire and perform write validation of data, wherein the system acquires and validates data generated by the one or more networked host nodes to perform secondary storage in a SAN-based data storage and recovery network;

(b) placing a metadata generator module in the system to passively acquire and perform write validation to generate metadata describing the data generated to store in primary storage, wherein the metadata generator is comprised of at least one of the one or more networked host nodes, a host node client, a data splitter, a line card, and a secondary storage server;

(c) coupling a primary SAN storage device to receive the data and metadata to the system to perform passive acquisition and write validation;

(d) forming a splitter to acquire passive access to data traveling a data path between a generating node and the primary SAN storage device and to mirror the data over an access line to a secondary storage server;

(e) placing the secondary storage server in the system to passively acquire and perform write validation to receive an exact copy of a data stream that passes the splitter; and

(f) forming a metadata comparison module to compare metadata associated with the data against additional metadata to validate or invalidate write data, wherein the additional metadata is stored in at least one of a local location and an external location, and wherein the metadata comparison module is comprised of at least one of the one or more networked host nodes, a host node client, a data splitter, a line card, and a secondary storage server.