Stub domain for request servicing

• US 9,448,827 B1
• Filed: 12/13/2013
• Issued: 09/20/2016
• Est. Priority Date: 12/13/2013
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Allen, Nicholas Alexander
• Primary Examiner(s): Puente, Emerson
• Assistant Examiner(s): Kamran, Mehran
• Application Number:
  US14/106,644
• Time in Patent Office: 1,012 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

Techniques for reclaiming resources from guest computing systems while those systems are waiting for responses to requests in virtualized and/or distributed computer systems are described herein. At a time after issuing a request and determining that the response will take longer than a threshold length of time, one or more computer system entities within a computer system invoke one or more computer system capabilities to at least instantiate a listener object, transfer the listener object to another system domain, suspend the guest computing system and reclaim resources from the suspended guest computing system. When the response is returned to the listener object, the guest computer system is restored and the response is forwarded to the restored guest. While the guest computing system is suspended, the reclaimed resources are made available to other computer system entities.

Claims

1. A computer-implemented method for suspending a guest computing system while waiting for a response to a request and for reclaiming computer system memory from the guest computing system while waiting for the response, comprising:

under the control of one or more computer systems configured with executable instructions,

issuing the request from the guest computing system to a second computing system that is configured to at least receive and respond to the request;

as a result of determining the response to the request will take a length of time greater than a threshold length of time, at least;

constructing a listener stub configured to at least receive the response and comprising at least;

a handling plan that at least includes a set of instructions for receiving the response; and

a listener socket configured to at least receive a network communication from the second computing system;

transferring the listener stub to a stub domain;

suspending the guest computing system; and

reclaiming at least a portion of the computer system memory from the guest computing system; and

as a result of the listener stub receiving the response from the second computing system, at least;

executing one or more of the instructions in the handling plan according to the response received;

restoring the guest computing system; and

forwarding the response from the stub domain to the guest computing system.

2. The computer-implemented method of claim 1, wherein the threshold length of time is based at least in part on a length of time to suspend and restore the guest computing system.

3. The computer-implemented method of claim 1, wherein the threshold length of time is based at least in part on an amount of available computer system memory on the guest computing system.

4. The computer-implemented method of claim 1, wherein the listener socket is created by the guest computing system and transferred to the stub domain as part of the listener stub.

5. The computer-implemented method of claim 1, wherein the listener socket is created by the stub domain based at least in part on one or more configuration parameters sent from the guest computing system to the stub domain as part of the listener stub.

6. The computer-implemented method of claim 1, wherein:

the listener stub further comprises a timer that is started when the listener stub is transferred to the stub domain; and

the stub domain is configured to respond to an expiration of the timer by at least;

restoring the guest computing system;

restoring the computer system memory to the guest computing system; and

forwarding a timeout response from the stub domain to the guest computing system.

7. A computer system, comprising:

one or more processors; and

memory including executable instructions that, when executed by the one or more processors, cause the computer system that is waiting for a response to a request issued by a guest computing system in a first domain of a virtualization environment, to at least;

as a result of determining the response to the request will take a length of time greater than a threshold length of time;

transfer a listener stub configured to at least receive the response to a stub domain of the virtualization environment, the listener stub comprising a listener socket;

suspend the guest computing system; and

reclaim resources from the guest computing system, the resources including computer system memory; and

as a result of the listener stub receiving at least a portion of the response, resulting in a received response, cause the listener stub to at least;

restore the guest computing system; and

forward the response to the guest computing system.

8. The computer system of claim 7, wherein the instructions further comprise instructions that, when executed by the one or more processors, cause the system, prior to constructing the listener stub, determine whether to construct the listener stub based at least in part on a determination that an expected length of time to obtain the response to the request exceeds the threshold length of time.

9. The computer system of claim 8, wherein the threshold length of time is based at least in part on one or more measurements of available system resources.

10. The computer system of claim 7, wherein the instructions further comprise instructions that, when executed by the one or more processors, cause the system, prior to constructing the listener stub, to determine whether to construct the listener stub based at least in part on a nomination sent from the guest computing system to a controlling domain wherein the nomination at least contains a notification to suspend the guest computing system.

11. The computer system of claim 7, wherein:

the listener socket is configured to at least receive the response over a network attached to the computer system; and

the listener stub comprises a handling plan at least comprising a set of instructions, one or more of which may be executed when the response is received.

12. The computer system of claim 7, wherein:

the stub domain is configured to at least;

receive the response; and

perform one or more actions as a result of receiving the response according to one or more sets of instructions for the listener stub.

13. The computer system of claim 7, wherein, to restore the guest computing system, the listener stub is configured to wait until the received response is completely received.

14. The computer system of claim 7, wherein the resources include at least computer system memory.

15. The computer system of claim 7, wherein forwarding the response to the guest computing system causes the computer system to at least:

create a duplicate copy of a network socket where the response is received; and

forward the duplicate copy of the network socket to the guest computing system.

16. The computer system of claim 7, wherein the response comprises a first portion that is received before the computer system is restored and a second portion that is received after the computer system is restored, and forwarding the response to the guest computing system causes the computer system to at least:

buffer the first portion of the response to result in a buffered first portion of the response;

forward the buffered first portion of the response to the guest computing system to result in a forwarded buffered first portion of the response; and

join the forwarded buffered first portion of the response with the second portion of the response.

17. A non-transitory computer-readable storage medium having collectively stored thereon executable instructions that, when executed by one or more processors of a computer system, cause the computer system to at least:

as a result of determining a response to a request will take a length of time greater than a threshold length of time, cause a guest computing system in first domain of a virtualization environment to wait for the response by at least;

transferring, to a stub domain of the virtualization environment, a listener stub configured to at least receive the response, the listener stub comprising a listener socket; and

reclaiming resources from the guest computing system, the resources comprising computer system memory; and

as a result of the listener stub receiving at least a portion of the response, cause the listener stub to process the response by;

restoring resources to the guest computing system; and

forwarding the response to the guest computing system.

18. The non-transitory computer-readable storage medium of claim 17, wherein the instructions further comprise instructions that, when executed by the one or more processes, cause the computer system to determine whether to implement the listener stub based at least in part on an expected amount of time to receive the response from the stub computing system.

19. The non-transitory computer-readable storage medium of claim 17, wherein the instructions further comprise instructions that, when executed by the one or more processes, cause the computer system to, prior to implementing the listener stub, determine whether to implement the listener stub based at least in part on the threshold length of time, wherein the threshold length of time is based at least in part on one or more measurements of availability of the resources.

20. The non-transitory computer-readable storage medium of claim 17, wherein the instructions that cause the computer system to reclaim resources from the guest computing system further include instructions that cause the computer system to suspend the guest computing system.

21. The non-transitory computer-readable storage medium of claim 17, wherein the instructions that cause the computer system to restore resources from the guest computing system further include instructions that cause the computer system to restore the guest computing system from a suspended state.

22. The non-transitory computer-readable storage medium of claim 17, wherein the instructions further comprise instructions that cause the computer system to:

receive the response; and

perform one or more actions as a result of receiving the response according to one or more sets of instructions in the listener stub.

23. The non-transitory computer-readable storage medium of claim 17, wherein the guest computing system resources include at least computer system memory.

24. The non-transitory computer-readable storage medium of claim 17, wherein the instructions that cause the computer system to forward the response to the guest computing system further include instructions that cause the computer system to:

buffer the portion of the response that is received before resources are restored to the guest computing system to result in a buffered portion of the response;

forward the buffered portion of the response to the guest computing system to result in a forwarded portion of the response; and

join the forwarded portion of the response with any portion of the response that is received after resources are restored to the guest computing system.