Local message queue processing for co-located workers

• US 8,954,993 B2
• Filed: 02/28/2013
• Issued: 02/10/2015
• Est. Priority Date: 02/28/2013
• Status: Expired due to Fees

• Current Assignee: Invincible IP LLC (IP Edge LLC)
• Original Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Inventors: Kruglick, Ezekiel
• Primary Examiner(s): Ho, Andy
• Application Number:
  US14/009,101
• Publication Number:
  US 20140245326A1
• Time in Patent Office: 712 Days
• Field of Search: 718/1, 719/313, 719/314
• US Class Current: 719/313
• CPC Class Codes: G06F 2209/548 Queue
• CPC Class Codes: G06F 9/546 Message passing systems or structures, e.g. queues
• CPC Class Codes: H04L 51/18 Commands or executable codes
• CPC Class Codes: H04L 51/214 using selective forwarding

Abstract

Technologies are provided for locally processing queue requests from co-located workers. In some examples, information about the usage of remote datacenter queues by co-located workers may be used to determine one or more matched queues. Messages from local workers to a remote datacenter queue classified as a matched queue may be stored locally. Subsequently, local workers that request messages from matched queues may be provided with the locally-stored messages.

Claims

1. A method to locally process queue requests from co-located workers in a datacenter, the method comprising:

detecting a producer worker at a first server sending a first message to a datacenter queue at least partially stored at a second server;

storing the first message in a queue cache at the first server, wherein the queue cache includes one of a copy and a partial copy of the datacenter queue;

detecting a consumer worker at the first server sending a message request to the datacenter queue;

providing the stored first message to the consumer worker in response to the message request;

receiving a signal from a command channel associated with the datacenter queue; and

modifying the stored first message in response to receiving the signal.

2. The method of claim 1, further comprising intercepting the first message sent by the producer worker before storing the first message.

3. The method of claim 1, wherein the producer worker and the consumer worker are co-located on a multi-core device at the first server.

4. The method of claim 1, wherein the producer worker and the consumer worker are executed on different virtual machines, the different virtual machines configured to execute on the same physical hardware.

5. The method of claim 1, wherein modifying the stored first message includes deleting the first message.

6. The method of claim 1, further comprising:

intercepting the message request from the consumer worker to the datacenter queue;

forwarding the message request to the datacenter queue if a first criterion is met; and

refraining from forwarding the message request to the datacenter queue if the first criterion is not met.

7. The method of claim 6, wherein the first criterion includes whether the datacenter queue is configured to hide a requested message upon receiving the message request from the consumer worker.

8. A virtual machine manager (VMM) to locally process queue requests from co-located workers in a datacenter, the VMM comprising:

a queue usage detector module configured to;

detect a producer worker at a first server, wherein the producer worker sends a first message to a datacenter queue at least partially stored at a second server; and

detect a consumer worker at the first server, wherein the consumer worker sends a message request to the datacenter queue, and wherein the producer worker and the consumer worker are co-located on a multi-core device at the first server; and

a processing module configured to;

intercept the first message sent by the producer worker;

store the first message at the first server;

provide the stored first message to the consumer worker in response to the message request;

receive a signal from a command channel associated with the datacenter queue; and

modify the stored first message in response to receiving the signal.

9. The VMM of claim 8, wherein the processing module is further configured to build a table of queue usage based on at least one observed datacenter queue request.

10. The VMM of claim 9, wherein the queue usage detector module is further configured to observe the at least one observed datacenter queue request.

11. The VMM of claim 8, wherein the producer worker and the consumer worker are executed on different virtual machines, the different virtual machines configured to execute on the same physical hardware.

12. The VMM of claim 8, wherein the processing module is further configured to:

intercept the message request from the consumer worker to the datacenter queue;

forward the message request to the datacenter queue if a first criterion is met; and

refrain from forwarding the message request to the datacenter queue if the first criterion is not met.

13. The VMM of claim 12, wherein the first criterion includes whether the datacenter queue is configured to hide the requested message upon receiving the message request from the consumer worker.

14. A cloud-based datacenter configured to locally process queue requests from co-located workers in the datacenter, the datacenter comprising:

a first and a second virtual machine (VM) operable to be executed on one or more physical machines; and

a datacenter controller configured to;

detect a producer worker that is executed on a first VM and sends a first message to a datacenter queue at least partially stored at a first datacenter location;

intercept the first message sent by the producer worker before storing the first message;

store the first message in a queue cache at a second datacenter location different from the first;

detect a consumer worker that is executed on a second VM and sends a message request to the datacenter queue;

provide the stored first message to the consumer worker in response to the message request, wherein the first message is stored and provided from within a server to the producer worker and the consumer worker;

receive a signal from a command channel associated with the datacenter queue; and

modify the stored first message in response to receiving the signal.

15. The datacenter of claim 14, wherein the controller is further configured to intercept the first message sent by the producer worker before storing the first message.

16. The datacenter of claim 14, wherein the queue cache includes one of a copy and a partial copy of the datacenter queue.

17. The datacenter of claim 14, wherein the controller is further configured to build a table of queue usage based on at least one observed datacenter queue request.

18. The datacenter of claim 17, wherein the controller is further configured to observe the at least one observed datacenter queue request.

19. The datacenter of claim 14, wherein the producer worker and the consumer worker are co-located on a multi-core device at the first datacenter location.

20. The datacenter of claim 14, wherein the first and second VMs are configured to execute on the same physical machine.

21. The datacenter of claim 14, wherein the controller is further configured to modify the stored first message by deleting the first message.

22. The datacenter of claim 14, wherein the controller is further configured to:

intercept the message request from the consumer worker to the datacenter queue;

forward the message request to the datacenter queue if a first criterion is met; and

refrain from forwarding the message request to the datacenter queue if the first criterion is not met.

23. The datacenter of claim 22, wherein the first criterion includes whether the datacenter queue is configured to hide the requested message upon receiving the message request from the consumer worker.