Granted Patent B1

US 7,460,528 · App. 10/414,891 · Granted Dec 2, 2008

Abstract

Claims (114)

1. A method for routing a data packet of an information unit sequence, comprising:

receiving at a switch a data packet of an information unit sequence of a block storage exchange from a storage client, the sequence associated with a source identifier and a target identifier identifying a target;

determining a storage resource identifier corresponding to the target identifier according to a forwarding table;

associating a sequence identifier with the source identifier, the sequence identifier corresponding to the target;

recording the source identifier and the sequence identifier at a context table;

routing the data packet according to the storage resource identifier;

receiving a next data packet associated with the sequence identifier; and

routing the next data packet according to the context table.

2. The method of claim 1 , wherein the sequence comprises a read command, and further comprising repeating the following until a final response is received from the storage resource:

receiving data from the storage resource; and

routing the data to the storage client using a hardware accelerator.

3. The method of claim 1 , wherein the sequence comprises a write command, and further comprising repeating the following until a final response is received from the storage client:

receiving data from the storage client; and

routing the data to the storage resource using a hardware accelerator.

4. The method of claim 1 , wherein the sequence comprises a write command corresponding to a plurality of storage resources, the plurality of storage resources comprising the storage resource, and further comprising repeating the following until a final response is received from the storage client:

receiving data from the storage client; and

routing the data to the plurality of storage resources using a hardware accelerator.

5. The method of claim 1 , wherein the sequence comprises a third party copy command to copy data from the storage resource to a destination resource, and further comprising repeating the following until a final response is received from the storage resource:

receiving data from the storage resource; and

routing the data to the destination resource.

6. A system for routing a data packet of an information unit sequence, comprising:

an interface of a switch operable to receive a data packet of an information unit sequence of a block storage exchange from a storage client, the sequence associated with a source identifier and a target identifier identifying a target; and

a processor coupled to the interface and operable to:

determine a storage resource identifier corresponding to the target identifier according to a forwarding table;

associate a sequence identifier with the source identifier the sequence identifier corresponding to the target;

record the source identifier and the sequence identifier at a context table;

route the data packet according to the storage resource identifier;

receive a next data packet associated with the sequence identifier; and

route the next data packet according to the context table.

7. The system of claim 6 , wherein the sequence comprises a read command, and the processor is further operable to repeat the following until a final response is received from the storage resource:

receive data from the storage resource: and

route the data to the storage client using a hardware accelerator.

8. The system of claim 6 , wherein the sequence comprises a write command, and the processor is further operable to repeat the following until a final response is received from the storage client:

receive data from the storage client; and

route the data to the storage resource using a hardware accelerator.

9. The system of claim 6 , wherein the sequence comprises a write command corresponding to a plurality of storage resources, the plurality of storage resources comprising the storage resource, the processor is further operable to repeat the following until a final response is received from the storage client:

receive data from the storage client; and

route the data to the plurality of storage resources using a hardware accelerator.

10. The system of claim 6 , wherein the sequence comprises a third party copy command to copy data from the storage resource to a destination resource, and the processor is further operable to repeat the following until a final response is received from the storage resource:

receive data from the storage resource; and

route the data to the destination resource.

11. Logic for routing a data packet of an information unit sequence, the logic embodied in a computer readable storage medium and operable to:

receive at a switch a data packet of an information unit sequence of a block storage exchange from a storage client, the sequence associated with a source identifier and a target identifier identifying a target;

determine a storage resource identifier corresponding to the target identifier according to a forwarding table;

associate a sequence identifier with the source identifier, the sequence identifier corresponding to the target;

record the source identifier and the sequence identifier at a context table;

route the data packet according to the storage resource identifier;

receive a next data packet associated with the sequence identifier; and

route the next data packet according to the context table.

12. The logic of claim 11 , wherein the sequence comprises a read command, and further operable to repeat the following until a final response is received from the storage resource:

receive data from the storage resource; and

route the data to the storage client using a hardware accelerator.

13. The logic of claim 11 , wherein the sequence comprises a write command, and further operable to repeat the following until a final response is received from the storage client:

receive data from the storage client; and

route the data to the storage resource using a hardware accelerator.

14. The logic of claim 11 , wherein the sequence comprises a write command corresponding to a plurality of storage resources, the plurality of storage resources comprising the storage resource, and further operable to repeat the following until a final response is received from the storage client:

receive data from the storage client; and

route the data to the plurality of storage resources using a hardware accelerator.

15. The logic of claim 11 , wherein the sequence comprises a third party copy command to copy data from the storage resource to a destination resource, and further operable to repeat the following until a final response is received from the storage resource:

receive data from the storage resource; and

route the data to the destination resource.

16. A system for routing a data packet of an information unit sequence, comprising:

means for receiving at a switch a data packet of an information unit sequence of a block storage exchange from a storage client, the sequence associated with a source identifier and a target identifier identifying a target;

means for determining a storage resource identifier corresponding to the target identifier according to a forwarding table;

means for associating a sequence identifier with the source identifier, the sequence identifier corresponding to the target;

means for recording the source identifier and the sequence identifier at a context table;

means for routing the data packet according to the storage resource identifier,

means for receiving a next data packet associated with the sequence identifier; and

routing the next data packet according to the context table.

17. A method for routing a data packet of an information unit sequence, comprising:

receiving at a switch a data packet of an information unit sequence of a block storage exchange from a storage client, the sequence associated with a source identifier and a target identifier identifying a target;

determining a storage resource identifier corresponding to the target identifier according to a forwarding table;

associating a sequence identifier with the source identifier, the sequence identifier corresponding to the target;

recording the source identifier and the sequence identifier at a context table;

routing the data packet according to the storage resource identifier;

receiving a next data packet associated with the sequence identifier, and routing the next data packet according to the context table;

repeating the following until a final response is received from the storage resource if the sequence comprises a read command: receiving data from the storage resource, and routing the data to the storage client using a hardware accelerator;

repeating the following until a final response is received from the storage client if the sequence comprises a first write command: receiving data from the storage client, and routing the data to the storage resource using a hardware accelerator;

repeating the following until a final response is received from the storage client if the sequence comprises a second write command corresponding to a plurality of storage resources, the plurality of storage resources comprising the storage resource: receiving data from the storage client, and routing the data to the plurality of storage resources using a hardware accelerator; and

repeating the following until a final response is received from the storage resource if the sequence comprises a third party copy command to copy data from the storage resource to a destination resource: receiving data from the storage resource, and routing the data to the destination resource.

18. A method for routing a data packet, comprising:

receiving at a switch a first data packet from a storage client, the first data packet having a first source address and a first destination address, the first destination address corresponding to a target identifier;

determining a storage resource address corresponding to the target identifier;

rewriting the first destination address of the first data packet at a hardware accelerator, the rewritten first destination address comprising the storage resource address;

routing the first data packet according to the rewritten first destination address;

assigning an identifier to the first data packet;

recording the first source address of the first data packet at the switch;

receiving at the switch a second data packet from a storage resource corresponding to the storage resource address, the second data packet having a second source address and a second destination address, the second data packet associated with the identifier assigned to the first data packet;

retrieving the recorded first source address in accordance with the identifier assigned to the first data packet;

rewriting the second destination address of the first data packet at the hardware accelerator, the rewritten second destination address comprising the first source address; and

routing the first data packet according to the rewritten second destination address.

19. A system for routing a data packet, comprising:

an input operable to receive at a switch a first data packet from a storage client, the first data packet having a first source address and a first destination address, the first destination address corresponding to a target identifier; and

a processor coupled to the input and operable to:

determine a storage resource address corresponding to the target identifier;

rewrite the first destination address of the first data packet at a hardware accelerator, the rewritten first destination address comprising the storage resource address;

route the first data packet according to the rewritten first destination address;

assign an identifier to the first data packet;

record the first source address of the first data packet at the switch;

receive at the switch a second data packet from a storage resource corresponding to the storage resource address, the second data packet having a second source address and a second destination address, the second data packet associated with the identifier assigned to the first data packet;

retrieve the recorded first source address in accordance with the identifier assigned to the first data packet;

rewrite the second destination address of the first data packet at the hardware accelerator, the rewritten second destination address comprising the first source address; and

route the first data packet according to the rewritten second destination address.

20. Logic for routing a data packet, the logic embodied in a computer readable storage medium and operable to:

receive at a switch a first data packet from a storage client, the first data packet having a first source address and a first destination address, the first destination address corresponding to a target identifier;

determine a storage resource address corresponding to the target identifier;

rewrite the first destination address of the first data packet at a hardware accelerator, the rewritten first destination address comprising the storage resource address;

route the first data packet according to the rewritten first destination address;

assign an identifier to the first data packet;

record the first source address of the first data packet at the switch;

receive at the switch a second data packet from a storage resource corresponding to the storage resource address, the second data packet having a second source address and a second destination address, the second data packet associated with the identifier assigned to the first data packet;

retrieve the recorded first source address in accordance with the identifier assigned to the first data packet;

rewrite the second destination address of the first data packet at the hardware accelerator, the rewritten second destination address comprising the first source address; and

route the first data packet according to the rewritten second destination address.

Assignments (8)