Dynamic allocation of resources while considering resource reservations

• US 10,063,491 B2
• Filed: 02/16/2017
• Issued: 08/28/2018
• Est. Priority Date: 06/21/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Das, Sudipto, Narasayya, Vivek, Li, Feng, Syamala, Manoj
• Primary Examiner(s): Boutah, Alina A
• Application Number:
  US15/435,269
• Publication Number:
  US 20170163554A1
• Time in Patent Office: 558 Days
• CPC Class Codes: H04L 47/72 using reservation actions during connection setup
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

Described herein are technologies relating to computing resource allocation among multiple tenants. Each tenant may have a respective absolute reservation for rate-based computing resources, which is independent of computing resource reservations of other tenants. The multiple tenants vie for the rate-based computing resources, and tasks are scheduled based upon which tenants submit the tasks and the resource reservations of such tenants.

Claims

1. A server computing device, comprising:

at least one processor core; and

memory that has a scheduler component loaded therein, wherein the scheduler component, when executed by the at least one processor core, is configured to;

maintain tasks that are executable by the server computing device, the tasks generated by tenants of the server computing device, the tenants having resource reservations for rate-based computing resources of the server computing device, wherein the rate-based computing resources are expressed as percentages of computing resources over defined time intervals; and

select a task from the tasks to be executed by the server computing device based upon the resource reservations of the tenants, the at least one processor core executes the task responsive to the task being selected by the scheduler component.

2. The server computing device of claim 1, wherein a first tenant in the tenants has a first resource reservation, the first resource reservation being a static reservation, the static reservation indicating that the first tenant has reserved a first threshold amount of rate-based computing resources of the server computing device for each metering interval.

3. The server computing device of claim 2, the static reservation indicating that the first tenant is not guaranteed access to rate-based computing resources of the server computing device above the first threshold amount of rate-based computing resources at any time.

4. The server computing device of claim 2, a second tenant in the tenants has a second resource reservation, the second resource reservation being an elastic reservation, the elastic reservation indicating that the second tenant has reserved a second threshold amount of rate-based computing resources of the server computing device over several metering intervals.

5. The server computing device of claim 4, the elastic reservation indicating that the second tenant is to have access to rate-based computing resources of the server computing device above the second threshold amount of rate-based computing resources when the second tenant has not utilized the second threshold amount of rate-based computing resources over the several metering intervals.

6. The server computing device of claim 4, the first threshold amount of rate-based computing resources and the second threshold amount of rate-based computing resources are expressed as one of percentages of fixed amount of rate-based computing resources of the server computing device, numbers of virtual cores, or CPU times.

7. The server computing device of claim 1, wherein the resource reservations are absolute and cumulatively greater than a fixed amount of computing resources of the server computing device.

8. The server computing device of claim 1, wherein a resource reservation in the resource reservations defines an amount of rate-based computing resources that is to be accessible to a tenant that has the resource reservation when the tenant requests rate-based computing resources of the at least one processor core.

9. The server computing device of claim 1, wherein the task is for a tenant in the tenants, and further wherein the scheduler component is configured to select the task based upon a penalty for failing to meet a resource reservation for the tenant.

10. The server computing device of claim 1, wherein each tenant is logically isolated computer-executable code.

11. A method executed by at least one processor core of a server computing device, the method comprising:

accessing a queue of tasks that are to be executed by the at least one processor core of the server computing device, wherein the tasks are input into the queue by tenants of the server computing device, the tenants have resource reservations for rate-based computing resources of the server computing device, wherein the resource reservations are expressed as percentages of the computing resources over defined windows of time;

selecting a task from the tasks to be executed by the at least one processor core, wherein the task is input to the queue by a tenant in the tenants, and further wherein the task is selected based upon the resource reservations of the tenants; and

executing the task from the tasks responsive to selecting the task.

12. The method of claim 11, wherein the task is selected based upon a penalty defined for the tenant that is applied when a resource reservation for the tenant is unmet.

13. The method of claim 11, wherein a first tenant in the tenants has a first resource reservation, the first resource reservation being a static reservation, the static reservation indicating that the first tenant has reserved a first threshold amount of rate-based computing resources of the server computing device for each metering interval.

14. The method of claim 13, wherein a second tenant in the tenants has a second resource reservation, the second resource reservation being an elastic reservation, the elastic reservation indicating that the second tenant has reserved a second threshold amount of rate-based computing resources of the server computing device over several metering intervals.

15. The method for claim 11, further comprising:

receiving computing resource utilizations of the tenants and the resource reservations of the tenants; and

computing fractional deficits for the tenants based upon the computing resource utilizations and the resource reservations, wherein the task is selected based upon the fractional deficits.

16. The method of claim 15, the computing resource utilizations of the tenants being with respect to metering intervals for the tenants, the metering intervals being time intervals during which the computing resource utilizations are computed and accounted.

17. The method of claim 15, wherein the task is selected based upon the tenant that input the task to the queue having a largest fractional deficit amongst the fractional deficits.

18. The method of claim 11, the task being a database task.

19. A computer-readable hardware storage medium comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:

receiving a plurality of tasks that are to be executed by a processor core, the plurality of tasks respectively correspond to a plurality of tenants, each tenant in the plurality of tenants being respective computer-executable code that generates tasks, the plurality of tenants have a respective plurality of resource reservations that include both static and elastic resource reservations, wherein the resource reservations are expressed as percentages of computing resources over defined periods of time;

selecting a task from the plurality of tasks based upon the plurality of resource reservations of the plurality of tenants; and

responsive to the task being selected, executing the task.

20. The computer-readable hardware storage medium of claim 19, wherein the resource reservations are absolute and cumulatively greater than an amount of computing resources of the processor.