Automatic determination of resource sizing

• US 9,537,788 B2
• Filed: 12/05/2014
• Issued: 01/03/2017
• Est. Priority Date: 12/05/2014
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Reque, Sean Philip, Manwaring, Derek Steven, Thomas, Dylan Chandler, Wagner, Timothy Allen, Zhao, Xin
• Primary Examiner(s): Dao, Thuy
• Assistant Examiner(s): Smith, Cheneca
• Application Number:
  US14/562,601
• Publication Number:
  US 20160164797A1
• Time in Patent Office: 760 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/45562 Creating, deleting, cloning virtual machine instances
• CPC Class Codes: G06F 9/445 Program loading or initiating bootstrapping G06F9/4401; security arrangements for program loading or initiating G06F21/57
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/808 User-type aware
• CPC Class Codes: H04L 47/83 based on usage prediction

Abstract

A system for providing automatic resource resizing is provided. The system may be configured to maintain a plurality of virtual machine instances. The system may be further configured to receive a request to execute a program code and allocate computing resources for executing the program code on one of the virtual machine instances. The amount of resources allocated for executing the program code may be specified by the request and adjusted as needed.

Claims

1. A computer-implemented method comprising:

as implemented by one or more computing devices configured with specific executable instructions,

maintaining a plurality of virtual machine instances on one or more physical computing devices;

determining, based on a request to execute a program code associated with a user on a virtual compute system, a user-specified amount of a first computing resource;

selecting one of the plurality of virtual machine instances to be used for executing the program code, the selected virtual machine instance having the first computing resource having a first maximum amount and a second computing resource having a second maximum amount, the second computing resource being different from the first computing resource;

determining a first ratio of the user-specified amount to the first maximum amount;

determining the corresponding amount of the second computing resource based on the determined first ratio and the second maximum amount, wherein a second ratio of the corresponding amount to the second maximum amount is identical to the first ratio;

selecting a first container having the user-specified amount of the first computing resource and the corresponding amount of the second computing resource for executing the program code;

causing the program code to be executed in the selected container;

recording actual amounts of the first and second computing resources used by the program code during the execution of the program code;

maintaining the user-specified amount in association with the program code associated with the user;

determining that the user-specified amount is to be overridden by an alternate amount specified in a subsequent request to execute the program code; and

selecting a second container having the alternate amount of the first or second computing resources for executing the program code.

2. The computer-implemented method of claim 1, further comprising recording errors or conditions detected during an execution of the program code, the errors or conditions being caused by a level of utilization of at least one of the first computing resource or the second computing resource.

3. The computer-implemented method of claim 1, further comprising over-subscribing the selected virtual machine instance such that a sum of the user-specified amount of the first computing resource of each container created in the selected virtual machine instance exceeds the first maximum amount.

4. The computer-implemented method of claim 1, further comprising maintaining a first class of users and a second class of users, wherein the first class of users are allowed to specify resource amounts for multiple computing resources, and the second class of users are restricted to specifying a resource amount for only one of the multiple computing resources.

5. A system adapted to provide low-latency computational capacity from a virtual compute fleet, the system comprising:

an electronic data store configured to store at least a program code of a user; and

a virtual compute system comprising one or more hardware computing devices adapted to execute specific computer-executable instructions, said virtual compute system in communication with the electronic data store, and configured to at least;

maintain a plurality of virtual machine instances on one or more physical computing devices, wherein the plurality of virtual machine instances comprise a warming pool comprising virtual machine instances having one or more software components loaded thereon and waiting to be assigned to a user, and an active pool comprising virtual machine instances currently assigned to one or more users;

receive a request to execute a program code associated with a user on the virtual compute system, the request including information indicating the program code and the user associated with the program code;

select from the warming pool or the active pool a virtual machine instance to execute the program code, the selected virtual machine instance having a first computing resource and a second computing resource, wherein a first maximum amount specifies a maximum amount of the first computing resource available to be provided by the selected virtual machine instance, and a second maximum amount specifies a maximum amount of the second computing resource available to be provided by the selected virtual machine instance;

determine, based on the received request, a user-specified amount of the first computing resource specified by the user;

determine a first ratio of the user-specified amount to the first maximum amount of the first computing resource;

determine a corresponding amount of the second computing resource based on the first ratio and the second maximum amount, wherein a second ratio of the corresponding amount to the second maximum amount is identical to the first ratio;

create a container in the selected virtual machine instance based on the user-specified amount of the first computing resource and the corresponding amount of the second computing resource; and

cause the program code associated with the user to be loaded from the electronic data store onto the container and executed in the container.

6. The system of claim 5, wherein the virtual compute system is further configured to record actual amounts of the first and second computing resources used for executing the program code in the container.

7. The system of claim 5, wherein the virtual compute system is further configured to over-subscribe the selected virtual machine instance such that a sum of the user-specified amount of the first or second computing resource of each container created in the selected virtual machine instance exceeds the first maximum amount.

8. A system, comprising:

a virtual compute system comprising one or more hardware computing devices executing specific computer-executable instructions and configured to at least;

maintain a plurality of virtual machine instances on one or more physical computing devices;

determine, based on a request to execute a program code associated with a user on the virtual compute system, a user-specified amount of a first computing resource;

select one of the plurality of virtual machine instances to be used for executing the program code, the selected virtual machine instance having the first computing resource having a first maximum amount and a second computing resource having a second maximum amount, the second computing resource being different from the first computing resource;

determine a first ratio of the user-specified amount to the first maximum amount;

determine the corresponding amount of the second computing resource based on the determined first ratio and the second maximum amount, wherein a second ratio of the corresponding amount to the second maximum amount is identical to the first ratio;

create a container in the selected virtual machine instance, the container having the user-specified amount of the first computing resource and the corresponding amount of the second computing resource;

cause the program code to be executed in the container;

record actual amounts of the first and second computing resources used by the program code during execution of the program code;

maintain the user-specified amount in association with the program code associated with the user;

determine that the user-specified amount is to be overridden by an alternate amount specified in a subsequent request to execute the program code;

allocate the alternate amount of the first or second computing resources to the subsequent request for executing the program code.

9. The system of claim 8, wherein the virtual compute system is further configured to record errors or conditions detected during an execution of the program code, the errors or conditions being caused by a level of utilization of at least one of the first computing resource or the second computing resource.

10. The system of claim 8, wherein the virtual compute system is further configured to maintain a first class of users and a second class of users, wherein the first class of users are allowed to specify resource amounts for multiple computing resources, and the second class of users are restricted to specifying a resource amount for only one of the multiple computing resources.

11. The system of claim 8, wherein the virtual compute system is further configured to determine, based on the user-specified amount, a resource amount of at least one computing resource other than the first and second computing resources.

12. The system of claim 8, wherein the virtual compute system is further configured to over-subscribe the selected virtual machine instance such that a sum of the user-specified amount of the first or second computing resource of each container created in the selected virtual machine instance exceeds the first maximum amount.

13. A computer-readable, non-transitory storage medium storing computer executable instructions that, when executed by one or more computing devices, configure the one or more computing devices to perform operations comprising:

maintaining a plurality of virtual machine instances on one or more physical computing devices;

determining, based on a request to execute a program code associated with a user on a virtual compute system, a user-specified amount of a first computing resource;

selecting one of the plurality of virtual machine instances to be used for executing the program code, the selected virtual machine instance having the first computing resource having a first maximum amount and the second computing resource having a second maximum amount, the second computing resource being different from the first computing resource;

determining a first ratio of the user-specified amount to the first maximum amount;

determining the corresponding amount of the second computing resource based on the determined first ratio and the second maximum amount, wherein a second ratio of the corresponding amount to the second maximum amount is identical to the first ratio;

selecting a first container having the user-specified amount of the first computing resource and the corresponding amount of the second computing resource to the request for executing the program code;

causing the program code to be executed in the selected container;

recording actual amounts of the first and second computing resources used by the program code during the execution of the program code;

maintaining the user-specified amount in association with the program code associated with the user;

determining that the user-specified amount is to be overridden by an alternate amount specified in a subsequent request to execute the program code; and

selecting a second container having the alternate amount of the first or second computing resources for executing the program code.

14. The computer-readable, non-transitory storage medium of claim 13, wherein the operations further comprise recording errors or conditions detected during an execution of the program code, the errors or conditions being caused by a level of utilization of at least one of the first computing resource or the second computing resource.

15. The computer-readable, non-transitory storage medium of claim 13, wherein the virtual compute system is further configured to over-subscribe the selected virtual machine instance such that a sum of the user-specified amount of the first or second computing resource of each container created in the selected virtual machine instance exceeds the first maximum amount.

16. The computer-readable, non-transitory storage medium of claim 13, wherein the operations further comprise maintaining a first class of users and a second class of users, wherein the first class of users are allowed to specify resource amounts for multiple computing resources, and the second class of users are restricted to specifying a resource amount for only one of the multiple computing resources.