Determining server utilization using upper bound values

Patent metadata

Field	Value
Publication number	US-11303550-B2
Application number	US-202017002048-A
Country	US
Kind code	B2
Filing date	Aug 25, 2020
Priority date	Jul 22, 2020
Publication date	Apr 12, 2022
Grant date	Apr 12, 2022

Abstract

Described embodiments provide systems and methods for monitoring server utilization and reallocating resources using upper bound values. A device can determine a value indicative of an upper bound of a processing load of a server using data points detected for the processing load over a first range of time. The upper bound can correspond to a percentage of the processing load during the first range of time. The device can monitor, using the value, the processing load of the server over a second range of time. A determination can be made whether the value of the processing load is greater than a threshold during the second range of time. The device can generate an alert for the device responsive to a comparison of the value of the processing load to the threshold.

First claim

We claim:

1. A method comprising:
determining, by a device, an upper bound of a processing load of a server using data points detected for the processing load over a first range of time, the upper bound determined based on a probability that the data points detected over the first range of time is less than or equal to the upper bound such that less than a total number of the data points over the first range of time are less than or equal to the upper bound;
determining, by the device, whether a value of a plurality of values of the upper bound of the processing load is greater than a threshold for processor utilization; and
generating, by the device, an alert responsive to a comparison of the value of the plurality of values of the upper bound of the processing load with the threshold for processor utilization.

2. The method of claim 1 , comprising:
generating, by the device, the alert responsive to the value of the plurality of values of the upper bound of the processing load being greater than the threshold at one or more points during the first range of time.

3. The method of claim 1 , comprising:
detecting, by the device, a degradation in a performance of the server responsive to the value of the plurality of values of the upper bound of the processing load being greater than the threshold at one or more points during the first range of time.

4. The method of claim 1 , comprising:
continuously determining, by the device, the value of the plurality of values of the upper bound of the processing load using a plurality of central processing unit (CPU) utilization loads for the server over a rolling window including the first range of time and a second range of time subsequent to the first range of time.

5. The method of claim 1 , comprising:
determining, by the device, a level component of the processing load and a residual component of the processing load.

6. The method of claim 1 , comprising:
applying, by the device, a filter to the processing load of the server to generate a moving average for the processing load; and
sampling, by the device using a determined window, the moving average of the processing load to generate a level component for the processing load, the determined window indicative of a plurality of portions of the first range of time.

7. The method of claim 1 , comprising:
subtracting, by the device, a moving average of the processing load from a signal indicative of the processing load to generate a residual component of the processing load, the signal indicative of original data points for the processing load of the server for the first range of time.

8. The method of claim 1 , comprising:
applying, by the device, a decomposition function to a level component of the processing load to generate a seasonal component for the processing load.

9. The method of claim 1 , comprising:
determining, by the device, a variance of a residual component of the processing load.

10. The method of claim 1 , comprising:
determining, by the device, the value of the plurality of values of the upper bound of the processing load of the server using a mean of an adjusted component of the processing load, a variance of the adjusted component of the processing load, a seasonal component of the processing load and a variance of a residual component of the processing load.

11. A system comprising:
one or more processors coupled to memory, the one or more processors configured to:
determine an upper bound of a processing load of a server using data points detected for the processing load over a first range of time, the upper bound determined based on a probability that the data points detected over the first range of time is less than or equal to the upper bound such that less than a total number of the data points over the first range of time are less than or equal to the upper bound;
determine whether a value of a plurality of values of the upper bound of the processing load is greater than a threshold for processor utilization; and
generate an alert responsive to a comparison of the value of the plurality of values of the upper bound of the processing load with the threshold for processor utilization.

12. The system of claim 11 , wherein the one or more processors are configured to:
generate the alert responsive to the value of the plurality of values of the upper bound of the processing load being greater than the threshold at one or more points during the first range of time.

13. The system of claim 11 , wherein the one or more processors are configured to:
detect a degradation in a performance of the server responsive to the value of the plurality of values of the upper bound of the processing load being greater than the threshold at one or more points during the first range of time.

14. The system of claim 11 , wherein the one or more processors are configured to:
continuously determine the value of the plurality of values of the upper bound of the processing load using a plurality of central processing unit (CPU) utilization loads for the server over a rolling window including the first range of time and a second range of time subsequent to the first range of time.

15. The system of claim 11 , wherein the one or more processors are configured to:
determine a level component of the processing load and a residual component of the processing load.

16. The system of claim 11 , wherein the one or more processors are configured to:
apply a filter to the processing load of the server to generate a moving average for the processing load; and
sample, using a determined window, the moving average of the processing load to generate a level component for the processing load, the determined window indicative of a plurality of portions of the first range of time.

17. The system of claim 11 , wherein the one or more processors are configured to:
subtract a moving average of the processing load from a signal indicative of the processing load to generate a residual component of the processing load, the signal indicative of original data points for the processing load of the server for the first range of time.

18. The system of claim 11 , wherein the one or more processors are configured to:
determine the value of the plurality of values of the upper bound of the processing load of the server using a mean of an adjusted component of the processing load, a variance of the adjusted component of the processing load, a seasonal component of the processing load and a variance of a residual component of the processing load.

19. A non-transitory computer-readable medium, comprising instructions that, when executed by the processor of a device, cause the processor to:
determine an upper bound of a processing load using data points detected for the processing load over a first range of time, the upper bound determined based on a probability that the data points detected over the first range of time is less than or equal to the upper bound such that less than a total number of the data points over the first range of time are less than or equal to the upper bound;
determine whether a value of a plurality of values of the upper bound of the processing load is greater than a threshold for processor utilization; and
generate an alert responsive to a comparison of the value of the plurality of values of the upper bound of the processing load to the threshold for processor utilization.

20. The computer-readable medium of claim 19 , further comprising instructions that cause the processor to:
generate t

Assignees

• Citrix Systems Inc

Inventors

• Varnavas Andreas
• Tiwari Satyendra
• Muthiah Manikam
• Georgakopoulos Nikolaos

Classifications

• G06F11/3433

  for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029) · CPC title

• G06F9/5083 Primary

  Techniques for rebalancing the load in a distributed system · CPC title

• H04L43/0817 Primary

  by checking functioning · CPC title

• H04L43/065

  related to network devices · CPC title

• G06F2209/508

  Monitor · CPC title