MULTICAST WRITE COMMANDS

• US 20130086320A1
• Filed: 09/30/2011
• Published: 04/04/2013
• Est. Priority Date: 09/30/2011
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Michael G. Myrah
• Inventors: Myrah, Michael G., Natrajan, Balaji, Black, Joseph David
• Granted Patent: US 8,909,864 B2
• Time in Patent Office: Days
• US Class Current: 711/118
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0683 Plurality of storage devices
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066

Abstract

Techniques for implementing a multicast write command are described. A data block may be destined for multiple targets. The targets may be included in a list. A multicast write command may include the list. Write commands may be sent to each target in the list.

Claims

1. A method comprising:

determining, at an initiator, if a block of data is to be written to more than one target;

selecting, with the initiator, a write manager;

generating, with the initiator, a multicast write command; and

sending the multicast write command from the initiator to the selected write manager.

2. The method of claim 1 wherein the write manager closest in fabric topology to the targets is selected.

3. The method of claim 1 wherein the multicast write command includes a list of targets to which the block of data will be written.

4. The method of claim 1 further comprising:

sending the block of data to the selected write manager.

5. The method of claim 1 further comprising:

performing a discovery process upon receipt of a change indication;

identifying accessible write managers in a fabric topology; and

storing locations within the fabric topology of the accessible write managers.

6. The method of claim 3 wherein the list of targets includes an offset for each target.

7. A non-transitory processor readable medium containing thereon a set of instructions which if executed by the processor cause the processor to:

receive a multicast write command from an initiator, the multicast write command including a list of targets;

send a write command to each target in the list of targets; and

send a ready response to the multicast write command to the initiator.

8. The non-transitory medium of claim 7 further comprising instructions which cause the processor to:

receive a data block from the initiator;

send the data block to each target in the list of targets;

receive responses from each target in the list of targets indicating successful receipt of the data block; and

sending a response to the initiator indicating the data block was successfully sent once responses from each target in the list of targets has been received.

9. The non-transitory medium of claim 7 further comprising instructions which cause the processor to:

respond to a discovery request by presenting the processor as a virtual PHY on a serial attached SCSI fabric.

10. The non-transitory medium of claim 8 further comprising instructions which cause the processor to:

cache the data block until the data block is sent to each target in the list of targets.

11. The non-transitory medium of claim 8 wherein the multicast write command includes an offset for each target in the list of targets, the offset indicating the location on each target wherein the data block is to be written.

12. A device comprising:

a plurality of ports, each port of the plurality of ports operatively coupled to a target;

a virtual PHY port operatively coupled to an initiator; and

a processor to receive a multicast write command including a list of targets from the initiator over the virtual PHY and to send write commands to each target in the list of targets over the plurality of ports.

13. The device of claim 12 wherein the processor further:

responds to a discovery process from the initiator with the address of the virtual PHY port.

14. The device of claim 12 wherein the processor further:

relays a data block between the virtual PHY port and the targets.

15. The device of claim 12 wherein the device is an expander on a serial attached SCSI fabric.