Prioritized throttling in a multiusage environment

• US 9,992,301 B2
• Filed: 06/26/2013
• Issued: 06/05/2018
• Est. Priority Date: 06/26/2013
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Sauerwald, Nikolai, Freidlin, David, Fischer, Andre
• Primary Examiner(s): JAKOVAC, RYAN J
• Application Number:
  US13/928,236
• Publication Number:
  US 20150006710A1
• Time in Patent Office: 1,805 Days
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/0888 Throughput
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network
• CPC Class Codes: H04L 67/61 taking into account QoS or priority requirements

Abstract

Prioritization criteria can be applied to assign a priority ranking to server processes executed by a backend server, and a utilization of the backend server by server demands can be quantified. Upon determining that the quantified utilization of the backend server has reached or exceeded one or more determined limits on backend utilization, at least a lowest priority server process can be throttled based on one or more throttling parameters.

Claims

1. A computer program product comprising a non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:

applying one or more prioritization criteria to assign a priority ranking to each of a plurality of server processes being executed by a backend server;

quantifying a utilization of the backend server by a plurality of server demands, the plurality of server demands comprising the plurality of server processes being executed by the backend server;

first determining that the quantified utilization of the backend server has reached or exceeded one or more determined limits on backend utilization;

throttling, in response to the first determining, at least a lowest priority server process of the plurality of server processes based on a first throttling parameter of one or more throttling parameters, wherein the throttling comprises allocating a first nonzero amount of one or more resources of the backend server to the lowest priority server process, wherein the first nonzero amount is smaller than a first amount of the one or more resources allocated to the lowest priority server process before throttling, and wherein the first throttling parameter is based on a utilization metric of the backend server with the lowest priority server process comprising a longest duration of utilization of the backend server;

assigning, in response to the throttling of at least the lowest priority server process, a different priority ranking to each of the plurality of server processes;

second determining, in response to the throttling of at least the lowest priority server process, whether a current utilization of the backend server is still at or above the one or more determined limits on backend utilization, wherein the different priority ranking is assigned based on whether a server process of the plurality of server processes has been throttled within a certain time period prior to the second determining; and

throttling, in response to the second determining and when the current utilization of the backend server is still at or above the one or more determined limits on backend utilization after the second determining, a next to lowest priority server process of the plurality of server processes by allocating a second nonzero amount of the one or more resources to the next to lowest priority server process, the throttling of the next to lowest priority server process based on a second throttling parameter of the one or more throttling parameters, wherein the second nonzero amount is smaller than a second amount of the one or more resources allocated to the next to lowest priority server process before throttling, wherein the next to lowest priority server process is determined based on the different priority ranking assigned to each of the plurality of server processes.

2. A computer program product as in claim 1, wherein the operations further comprise:

determining, in response to the throttling of at least the lowest priority server process and/or the throttling of the next to lowest priority server process, an amount a maximum possible load on the backend server can be increased; and

de-throttling, based on the determined amount and when the quantified utilization satisfies a de-throttling threshold, a highest priority, previously throttled server process.

3. A computer program product as in claim 1, wherein the prioritization criteria comprise at least one of a program type, a specific program, a specific user or a group of users, a business case or group of business cases.

4. A computer program product as in claim 1, wherein the quantifying further comprises at least one of determining a percentage of available processor cycles currently in use, determining a number of calls to the backend server per a given unit of time, and using one or more key performance indicators.

5. A computer program product as in claim 1, wherein the one or more throttling parameters are defined universally and/or with one or more aspects that are specific to at least one of a specific server process, a type of server process, a specific user, and a specific type of user.

6. A computer program product as in claim 1, wherein the one or more throttling parameters comprise a plurality of sets of throttling parameters that are applied stepwise as the utilization of the backend server increases.

7. A system comprising:

at least one programmable processor; and

a machine-readable medium storing instructions that, when executed by the at least one programmable processor, cause the at least one programmable processor to perform operations comprising;

applying one or more prioritization criteria to assign a priority ranking to each of a plurality of server processes being executed by a backend server;

quantifying a utilization of the backend server by a plurality of server demands, the plurality of server demands comprising the plurality of server processes being executed by the backend server;

first determining that the quantified utilization of the backend server has reached or exceeded one or more determined limits on backend utilization;

throttling, in response to the first determining, at least a lowest priority server process of the plurality of server processes based on a first throttling parameter of one or more throttling parameters, wherein the throttling comprises allocating a first nonzero amount of one or more resources of the backend server to the lowest priority server process, wherein the first nonzero amount is smaller than a first amount of the one or more resources allocated to the lowest priority server process before throttling, and wherein the first throttling parameter is based on a utilization metric of the backend server with the lowest priority server process comprising a longest duration of utilization of the backend server;

assigning, in response to the throttling of at least the lowest priority server process, a different priority ranking to each of the plurality of server processes;

second determining, in response to the throttling of at least the lowest priority server process, whether a current utilization of the backend server is still at or above the one or more determined limits on backend utilization, wherein the different priority ranking is assigned based on whether a server process of the plurality of server processes has been throttled within a certain time period prior to the second determining;

throttling, in response to the second determining and when the current utilization of the backend server is still at or above the one or more determined limits on backend utilization after the second determining, a next to lowest priority server process of the plurality of server processes by allocating a second nonzero amount of the one or more resources to the next to lowest priority server process, the throttling of the next to lowest priority server process based on a second throttling parameter of the one or more throttling parameters, wherein the second nonzero amount is smaller than a second amount of the one or more resources allocated to the next to lowest priority server process before throttling, wherein the next to lowest priority server process is determined based on the different priority ranking assigned to each of the plurality of server processes;

third determining, in response to the throttling of at least the lowest priority server process and/or the throttling of the next to lowest priority server process, an amount a maximum possible load on the backend server can be increased and that the quantified utilization of the backend server satisfies a de-throttling threshold; and

de-throttling, in response to the third determining, a highest priority, previously throttled server process.

8. A system as in claim 7, wherein the de-throttling threshold is less than the one or more determined limits on backend utilization.

9. A system as in claim 7, wherein the prioritization criteria comprise at least one of a program type, a specific program, a specific user or a group of users, a business case or group of business cases.

10. A system as in claim 7, wherein the quantifying further comprises at least one of determining a percentage of available processor cycles currently in use, determining a number of calls to the backend server per a given unit of time, and using one or more key performance indicators.

11. A system as in claim 7, wherein the one or more throttling parameters are defined universally and/or with one or more aspects that are specific to at least one of a specific server process, a type of server process, a specific user, and a specific type of user.

12. A system as in claim 7, wherein the one or more throttling parameters comprise a plurality of sets of throttling parameters that are applied stepwise as the utilization of the backend server increases.

13. A computer-implemented method comprising:

applying one or more prioritization criteria to assign a priority ranking to each of a plurality of server processes being executed by a backend server;

quantifying a utilization of the backend server by a plurality of server demands, the plurality of server demands comprising the plurality of server processes being executed by the backend server;

first determining that the quantified utilization of the backend server has reached or exceeded one or more determined limits on backend utilization;

throttling, in response to the first determining, at least a lowest priority server process of the plurality of server processes based on a first throttling parameter of one or more throttling parameters, wherein the throttling comprises allocating a first nonzero amount of one or more resources of the backend server to the lowest priority server process, wherein the first nonzero amount is smaller than a first amount of the one or more resources allocated to the lowest priority server process before throttling, and wherein the first throttling parameter is based on a utilization metric of the backend server with the lowest priority server process comprising a longest duration of utilization of the backend server;

assigning, in response to the throttling of at least the lowest priority server process, a different priority ranking to each of the plurality of server processes;

second determining, in response to the throttling of at least the lowest priority server process, whether a current utilization of the backend server is still at or above the one or more determined limits on backend utilization, wherein the different priority ranking is assigned based on whether a server process of the plurality of server processes has been throttled within a certain time period prior to the second determining; and

throttling, in response to the second determining and when the current utilization of the backend server is still at or above the one or more determined limits on backend utilization after the second determining, a next to lowest priority server process of the plurality of server processes by allocating a second nonzero amount of the one or more resources to the next to lowest priority server process, the throttling of the next to lowest priority server process based on a second throttling parameter of the one or more throttling parameters, wherein the second nonzero amount is smaller than a second amount of the one or more resources allocated to the next to lowest priority server process before throttling, wherein the next to lowest priority server process is determined based on the different priority ranking assigned to each of the plurality of server processes.

14. A computer-implemented method as in claim 13, further comprising:

determining, in response to the throttling of at least the lowest priority server process and/or the throttling of the next to lowest priority server process, an amount a maximum possible load on the backend server can be increased; and

de-throttling, based on the determined amount and when the quantified utilization satisfies a de-throttling threshold, a highest priority, previously throttled server process.

15. A computer-implemented method as in claim 13, wherein the prioritization criteria comprise at least one of a program type, a specific program, a specific user or a group of users, a business case or group of business cases.

16. A computer-implemented method as in claim 13, wherein the quantifying further comprises at least one of determining a percentage of available processor cycles currently in use, determining a number of calls to the backend server per a given unit of time, and using one or more key performance indicators.

17. A computer-implemented method as in claim 13, wherein the one or more throttling parameters are defined universally and/or with one or more aspects that are specific to at least one of a specific server process, a type of server process, a specific user, and a specific type of user.

18. A computer-implemented method as in claim 13, wherein the one or more throttling parameters comprise a plurality of sets of throttling parameters that are applied stepwise as the utilization of the backend server increases.

19. A computer-implemented method as in claim 13, wherein at least one of the applying, the quantifying, the determining, and the throttling is performed by a system comprising at least one programmable processor.

20. A computer program product comprising a non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:

applying one or more prioritization criteria to assign a priority ranking to each of a plurality of server processes being executed by a backend server, the plurality of server processes prioritized to include a lowest priority server process and a next to lowest priority server process;

first determining that a quantified utilization of the backend server has reached or exceeded one or more limits on backend utilization;

first throttling, when the quantified utilization of the backend server is determined to have reached or exceeded one or more limits on backend utilization, the lowest priority server process by at least allocating a first nonzero amount of one or more resources of the backend server to the lowest priority server process, wherein the first nonzero amount is smaller than a first amount of the one or more resources allocated to the lowest priority server process before the first throttling;

assigning, in response to the throttling of at least the lowest priority server process, a different priority ranking to each of the plurality of server processes;

second determining, in response to the throttling of at least the lowest priority server process, whether a current utilization of the backend server is still at or above the one or more determined limits on backend utilization, the different priority ranking assigned based on whether a server process of the plurality of server processes has been throttled within a certain time period prior to the second determining; and

throttling, in response to the second determining and when the current utilization of the backend server is still at or above the one or more determined limits on backend utilization after the second determining, a next to lowest priority server process of the plurality of server processes by allocating a second nonzero amount of the one or more resources to the next to lowest priority server process, the throttling of the next to lowest priority server process based on a second throttling parameter of the one or more throttling parameters, wherein the second nonzero amount is smaller than a second amount of the one or more resources allocated to the next to lowest priority server process before throttling, wherein the next to lowest priority server process is determined based on the different priority ranking assigned to each of the plurality of server processes.