Granted Patent B2

US 9,052,953 · App. 13/478,710 · Granted Jun 9, 2015

Inventor: Juliano Maldaner (Cupertino, CA)

Assignee: Citrix Systems, Inc.

G06F9/5011H04L43/0876H04L43/16G06F9/5077

Abstract

Claims (86)

1. A method comprising:

storing a capacity estimation value in computer memory, wherein the capacity estimation value estimates how many concurrent sessions a resource in a virtualized computing environment has workload capacity to handle;

monitoring, by a processor, a plurality of resource consumption metrics of the resource on an ongoing basis, wherein the plurality of resource consumption metrics comprise a network bandwidth consumption metric, CPU consumption metric, and memory consumption metric;

updating the capacity estimation value using the plurality of monitored resource consumption metrics;

comparing, by the processor, the capacity estimation value in response to receiving a request to create a new session in the resource;

receiving a hardware parameter of the resource and the plurality of resource consumption metric values of the resource;

determining a current number of sessions being handled by the resource;

recording, in computer memory, a plurality of utilization measurements over time using at least the resource consumption metric value and a corresponding current number of sessions;

determining that the recorded measurements fall within a predetermined deviation margin;

averaging the recorded measurements to calculate the capacity estimation value; and

updating the capacity estimation value using the calculated capacity estimation value.

2. The method of claim 1 , wherein updating the capacity estimation value includes:

increasing the capacity estimation value if the monitoring indicates that the resource has workload capacity to support more total number of concurrent sessions, and

decreasing the capacity estimation value if the monitoring indicates that the resource has workload capacity to support less total number of concurrent sessions.

3. The method of claim 1 , comprising:

receiving a hardware parameter of a TYPE 1 component included in the resource;

receiving resource consumption metric value of a TYPE 1 component included in the resource;

determining a current number of sessions being handled by the resource;

recording, in computer memory, a plurality of utilization measurements over time using at least the resource consumption metric value and the corresponding current number of sessions;

extrapolating the recorded utilization measurements to calculate the capacity estimation value; and

updating the capacity estimation value with the calculated value.

4. The method of claim 3 , comprising:

determining that the utilization measurements fall within a predetermined deviation margin, before the updating of the capacity estimation value.

5. The method of claim 1 , comprising:

receiving notification of an added computing device to the resource in the virtualized computing environment;

determining an estimated workload capacity of the added computing device;

updating the capacity estimation value using the estimated workload capacity of the added computing device.

6. The method of claim 5 , wherein the estimated workload capacity of the added computing device is calculated using a current average for the resource extrapolated based on the added computing device's hardware parameter.

7. The method of claim 5 , wherein the updating the capacity estimation value after the determining the estimated workload capacity of the added computing device comprises:

Calculating a weighted average of the initial capacity estimation value and the updated capacity estimation value; and

updating the capacity estimation value with the calculated weighted average.

8. The method of claim 1 , wherein the resource includes TYPE 1 and TYPE 2 components, and a first monitored resource consumption metric corresponds to a TYPE 1 component and a second monitored resource consumption metric corresponds to a TYPE 2 component.

9. The method of claim 1 , wherein the new session is a long-lived session.

10. The method of claim 1 , comprising:

receiving a request from a remote computing device to create the new session in the resource in the virtualized computing environment;

determining that an actual consumed capacity of the resource in the virtualized computing environment is less than the capacity estimation value for the resource; and

creating the new session between the resource in the virtualized computing environment and the remote computing device;

receiving a plurality of requests over the new session to perform computations over a period of time; and

closing the new session.

11. The method of claim 1 , comprising:

before the requested new session is created, determining that the resource in the virtualized computing environment and the requested new session are configured to handle the workload of a first group, wherein the first group comprises user sessions with similar utilization measurements.

12. The method of claim 1 , comprising:

rejecting the received request to create the new session if the request is associated with a user session associated with a group that is not handled by the resource in the virtualized computing environment; and

rejecting the received request to create the new session if the comparing indicates that the capacity estimation would be exceeded.

13. An apparatus, comprising:

at least one processor configured to execute computer-executable instructions; and

at least one memory storing the computer-executable instructions, which when executed by the processor, cause the apparatus to:

store a capacity estimation value, wherein the capacity estimation value estimates workload capacity of a resource in a virtualized computing environment;

monitor a plurality of resource consumption metrics of the resource on an ongoing basis, wherein the plurality of resource consumption metrics comprise a network bandwidth consumption metric, CPU consumption metric, and memory consumption metric;

update the capacity estimation value using the plurality of monitored resource consumption metrics;

compare the capacity estimation value in response to receiving a request to create a new session in the resource;

reject the received request to create the new session if the comparing indicates that the capacity estimation would be exceeded;

receive a hardware parameter of the resource and the plurality of resource consumption metric values of the resource;

determine a current number of sessions being handled by the resource;

record a plurality of utilization measurements over time using at least the resource consumption metric value and a corresponding current number of sessions;

determine that the recorded measurements fall within a predetermined deviation margin;

average the recorded measurements to calculate the capacity estimation value; and

update the capacity estimation value using the calculated capacity estimation value.

14. The apparatus of claim 13 , wherein the new session is a long-lived session.

15. The apparatus of claim 13 , wherein the virtualized computing environment comprises a hypervisor, a plurality of virtual machines, and a control program.

16. The apparatus of claim 13 , wherein the memory further stores computer-executable instructions, which when executed by the processor, cause the apparatus to:

before creating requested new session, determine that the resource in the virtualized computing environment and the requested new session are configured to handle the workload of a first group, wherein the first group comprises user sessions with similar typical utilization measurements; and

reject the received request to create the new session if the request is not associated with a user session associated with the first group.

17. A method comprising:

storing a capacity estimation value in computer memory, wherein the capacity estimation value estimates workload capacity of a resource in a virtualized computing environment;

monitoring, by a processor, at least one resource consumption metric of the resource on an ongoing basis, wherein the at least one resource consumption metric comprises CPU consumption metric;

updating the capacity estimation value using the monitored at least one resource consumption metric;

comparing, by the processor, the capacity estimation value in response to receiving a request to create a new session in the resource;

receiving notification of an added computing device to the resource in the virtualized computing environment;

determining an estimated workload capacity of the added computing device; and

updating the capacity estimation value using the estimated workload capacity of the added computing device, comprising:

calculating a weighted average of an initial capacity estimation value and the updated capacity estimation value; and

updating the capacity estimation value with the calculated weighted average.

18. The method of claim 17 , wherein the at least one resource consumption metric further comprises a network bandwidth consumption metric and memory consumption metric, and the method further comprising:

rejecting the received request to create the new session if the request is associated with a user session associated with a group that is not handled by the resource in the virtualized computing environment; and

rejecting the received request to create the new session if the comparing indicates that the capacity estimation would be exceeded.

19. An apparatus, comprising:

at least one processor configured to execute computer-executable instructions; and

at least one memory storing the computer-executable instructions that, when executed by the at least one processor, cause the apparatus to:

store a capacity estimation value in the at least one memory, wherein the capacity estimation value estimates workload capacity of a resource in a virtualized computing environment;

monitor at least one resource consumption metric of the resource on a periodic basis, wherein the at least one resource consumption metric comprises CPU consumption metric;

update the capacity estimation value using the monitored at least one resource consumption metric;

receive notification of an added computing device to the resource in the virtualized computing environment;

determine an estimated workload capacity of the added computing device;

calculate a weighted average of an initial capacity estimation value and the capacity estimation value updated using the estimated workload capacity of the added computing device; and

update the capacity estimation value with the calculated weighted average.

Assignments (9)

Continuity (2)