System and method for managing system resources in a network environment

• US 7,930,395 B2
• Filed: 12/01/2008
• Issued: 04/19/2011
• Est. Priority Date: 12/01/2008
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Moribayashi, Akira
• Primary Examiner(s): SALAD, ABDULLAHI ELMI
• Application Number:
  US12/325,326
• Publication Number:
  US 20100138510A1
• Time in Patent Office: 869 Days
• Field of Search: 709223-226
• US Class Current: 709/224
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]

Abstract

A method for managing system resources in a network environment. The method begins by reading from a first log both timestamp T_i and actual response R_i for i=1, 2, . . . , I. Each T_i comprising a timestamp corresponding to a request to utilize system resources. Each R_i comprising a length of time necessary for the system to respond to the request. After reading timestamps and response times, for each combination of both T_i and R_i where R_i does not equal zero: calculating active thread AT(T_i+j) for j=0, 1, . . . , R_i−1, according to a function AT(T_i+j)=AT(T_i+j)+1. Active thread AT(T_i+j) comprising the total number of threads active in the system at time T_i+j. Finally, storing active thread AT(T_i+j) for in a second log. The second log comprising both timestamp T_i+j and active thread AT(T_i+j).

Claims

1. A method for managing system resources in a network environment, said method comprising:

reading from a first log both timestamp T_i and actual response R_i for i=1, 2, . . . , I, said T_i comprising a timestamp corresponding to a request to utilize said system resources, said R_i comprising a length of time necessary for said system to respond to said request, said R_i being expressed in dimensionless units, said I corresponding to the total number of both said timestamps and said actual responses residing in said first log, said I being a positive integer having a value greater than 1; and

for each combination of both timestamp T_i and actual response R_i for i=1, 2, . . . , I, where R_i does not equal zero;

calculating a number active threads AT(T_i+j) in said system at time T_i+j for j=0, 1, . . . , R₁−

1, according to a function AT(T_i+j)=AT(T_i+j)+1, said AT(T_i+j) comprising the total number of threads active at time T_i+j; and

storing AT(T_i+j) for j=0, 1, . . . , R_i−

1, in a second log, said second log comprising both timestamp T_i+j and AT(T_i+j).

2. The method of claim 1, said method further comprising:

comparing a threshold value to each said active thread AT_y for y=1, 2, . . . , Y, said index Y being a positive integer equal to the total number of active thread values residing in said second log, said threshold value being a positive integer having a value greater than 1; and

sending a message to an output device if said comparing determines that at least one said active thread AT_y has a value greater than said threshold value.

3. The method of claim 2, wherein said threshold value is both provided by an administrator and received prior to said comparing.

4. The method of claim 2, said message further comprising:

each of said at least one active thread AT_y determined to have a value greater than said threshold value; and

each timestamp T_y corresponding to each said at least one active thread AT_y determined to have a value greater than said threshold value.

5. The method of claim 1, wherein prior to said reading from said first log:

reading from a third log timestamp T_i for i=1, 2, . . . , I;

after said reading from said third log, calculating estimated response E_i for i=1, 2, . . . , I, according to a function E_i=max(T_x|1<

x<

i);

calculating said actual response R_i for i=1, 2, . . . , I, according to a function R_i=E_i−

T_i, wherein if said R_i is equal to a negative number then R_i is changed to being equal to 0; and

storing said actual response Ri for i=1, 2, . . . , I, in said first log, said second log comprising both timestamp T_i and actual response R_i for i=1, 2, . . . , I.

6. The method of claim 5, wherein prior to said reading from said third log:

monitoring a network server for at least one said request to utilize said system resources;

monitoring said network server for at least one response, each response of said at least one response corresponding to a unique request of said at least one said request;

in response to having identified said at least one said request, recording a timestamp T corresponding to the time in which said at least one said request was identified; and

in response to having identified said at least one said response, storing said timestamp T in said third log.

7. A computer program product, comprising a computer readable storage device having a computer-readable program code stored therein, said computer-readable program code containing instructions that when executed by a processor of a computer system implement a method for managing system resources in a network environment, said method comprising:

reading from a first log both timestamp T_i and actual response R_i for i=1, 2, . . . , I, said T_i comprising a timestamp corresponding to a request to utilize said system resources, said R_i comprising a length of time necessary for said system to respond to said request, said R_i being expressed in dimensionless units, said I corresponding to the total number of both said timestamps and said actual responses residing in said first log, said I being a positive integer having a value greater than 1; and

for each combination of both timestamp T_i and actual response R_i for i=1, 2, . . . , I, where R_i does not equal zero;

calculating a number active threads AT(T_i+j) in said system at time T_i+j for j=0, 1, . . . , R_i−

1, according to a function AT(T_i+j)=AT(T_i+j)+1, said AT(T_i+j) comprising the total number of threads active at time T_i+j; and

storing AT(T_i+j) for j=0, 1, . . . , R_i−

1, in a second log, said second log comprising both timestamp T_i+j and AT(T_i+j).

8. The computer program product of claim 7, said method further comprising:

comparing a threshold value to each said active thread AT_y for y=1, 2, . . . , Y, said index Y being a positive integer equal to the total number of active thread values residing in said second log, said threshold value being a positive integer having a value greater than 1; and

sending a message to an output device if said comparing determines that at least one said active thread AT_y has a value greater than said threshold value.

9. The computer program product of claim 8, wherein said threshold value is both provided by an administrator and received prior to said comparing.

10. The computer program product of claim 8, said message further comprising:

each of said at least one active thread AT_y determined to have a value greater than said threshold value; and

each timestamp T_y corresponding to each said at least one active thread AT_y determined to have a value greater than said threshold value.

11. The computer program product of claim 7, wherein prior to said reading from said first log:

reading from a third log timestamp T_i for i=1, 2, . . . , I;

after said reading from said third log, calculating estimated response E_i for i=1, 2, . . . , I, according to a function E_i=max(T_x|1<

x<

i);

calculating said actual response R_i for i=1, 2, . . . , I, according to a function R_i=E_i−

T_i, wherein if said R_i is equal to a negative number then R_i is changed to being equal to 0; and

storing said actual response Ri for i=1, 2, . . . , I, in said first log, said second log comprising both timestamp T_i and actual response R_i for i=1, 2, . . . , I.

12. The computer program product of claim 11, wherein prior to said reading from said third log:

monitoring a network server for at least one said request to utilize said system resources;

monitoring said network server for at least one response, each response of said at least one response corresponding to a unique request of said at least one said request;

in response to having identified said at least one said request, recording a timestamp T corresponding to the time in which said at least one said request was identified; and

in response to having identified said at least one said response, storing said timestamp T in said third log.

13. A computing system comprising a processor coupled to a computer-readable memory unit, said memory unit comprising a software application, said software application comprising instruction that when executed by said processor, implement a method for managing system resources in a network environment, said method comprising:

reading from a first log both timestamp T_i and actual response R_i for i=1, 2, . . . , I, said T_i comprising a timestamp corresponding to a request to utilize said system resources, said R_i comprising a length of time necessary for said system to respond to said request, said R_i being expressed in dimensionless units, said I corresponding to the total number of both said timestamps and said actual responses residing in said first log, said I being a positive integer having a value greater than 1; and

for each combination of both timestamp T_i and actual response R_i for i=1, 2, . . . , I, where does not equal zero;

calculating a number active threads AT(T_i+j) in said system at time T_i+j for j=0, 1, . . . , R_i−

1, according to a function AT(T_i+j)=AT(T_i+j)+1, said AT(T_i+j) comprising the total number of threads active at time T_i+j; and

storing AT(T_i+j) for j=0, 1, . . . , R_i−

1, in a second log, said second log comprising both timestamp T_i+j and AT(T_i+j).

14. The computing system of claim 13, said method further comprising:

comparing a threshold value to each said active thread AT_y for y=1, 2, . . . , Y, said index Y being a positive integer equal to the total number of active thread values residing in said second log, said threshold value being a positive integer having a value greater than 1; and

sending a message to an output device if said comparing determines that at least one said active thread AT_y has a value greater than said threshold value.

15. The computing system of claim 14, wherein said threshold value is both provided by an administrator and received prior to said comparing.

16. The computing system of claim 14, said message further comprising:

each of said at least one active thread AT_y determined to have a value greater than said threshold value; and

each timestamp T_y corresponding to each said at least one active thread AT_y determined to have a value greater than said threshold value.

17. The computing system of claim 13, wherein prior to said reading from said first log:

reading from a third log timestamp T_i for i=1, 2, . . . , I;

after said reading from said third log, calculating estimated response E_i for i=1, 2, . . . , I, according to a function E_i=max(T_x|1<

x<

i);

calculating said actual response for i=1, 2, . . . , I, according to a function R_i=E_i−

T_i, wherein if said is equal to a negative number then R_i is changed to being equal to 0; and

storing said actual response Ri for i=1, 2, . . . , I, in said first log, said second log comprising both timestamp T_i and actual response R_i for i=1, 2, . . . , I.

18. The computing system of claim 17, wherein prior to said reading from said third log:

monitoring a network server for at least one said request to utilize said system resources;

monitoring said network server for at least one response, each response of said at least one response corresponding to a unique request of said at least one said request;

in response to having identified said at least one said request, recording a timestamp T corresponding to the time in which said at least one said request was identified; and

in response to having identified said at least one said response, storing said timestamp T in said third log.

19. A process for supporting computer infrastructure, said process comprising providing at least one support service for at least one of creating, integrating, hosting, maintaining, and deploying computer-readable code in a computing system, wherein said code in combination with the computing system is capable of implementing a method for managing system resources in a network environment, said method comprising:

reading from a first log both timestamp T_i and actual response R_i for i=1, 2, . . . , I, said T_i comprising a timestamp corresponding to a request to utilize said system resources, said R_i comprising a length of time necessary for said system to respond to said request, said R_i being expressed in dimensionless units, said I corresponding to the total number of both said timestamps and said actual responses residing in said first log, said I being a positive integer having a value greater than 1; and

for each combination of both timestamp T_i and actual response R_i for i=1, 2, . . . , I, where R_i does not equal zero;

calculating a number active threads AT(T_i+j) in said system at time T_i+j for j=0, 1, . . . , R_i−

1, according to a function AT(T_i+j_j)=AT(T_i+j)+1, said AT(T_i+j) comprising the total number of threads active at time T_i+j; and

storing AT(T_i+j) for j=0, 1, . . . , R_i−

1, in a second log, said second log comprising both timestamp T_i+j and AT(T_i+j).

20. The process for supporting computer infrastructure of claim 19, said method further comprising:

comparing a threshold value to each said active thread AT_y for y=1, 2, . . . , Y, said index Y being a positive integer equal to the total number of active thread values residing in said second log, said threshold value being a positive integer having a value greater than 1; and

sending a message to an output device if said comparing determines that at least one said active thread AT_y has a value greater than said threshold value.

21. The process for supporting computer infrastructure of claim 20, wherein said threshold value is both provided by an administrator and received prior to said comparing.

22. The process for supporting computer infrastructure of claim 20, said message further comprising:

each of said at least one active thread AT_y determined to have a value greater than said threshold value; and

each timestamp T_y corresponding to each said at least one active thread AT_y determined to have a value greater than said threshold value.

23. The process for supporting computer infrastructure of claim 19, wherein prior to said reading from said first log:

reading from a third log timestamp T_i for i=1, 2, . . . , I;

after said reading from said third log, calculating estimated response E_i for i=1, 2, . . . , I, according to a function E_i=max(T_x|1<

x<

i);

calculating said actual response R_i for i=1, 2, . . . , I, according to a function R_i=E_i−

T_i, wherein if said R_i is equal to a negative number then R_i is changed to being equal to 0; and

storing said actual response Ri for i=1, 2, . . . , I, in said first log, said second log comprising both timestamp T_i and actual response R_i for i=1, 2, . . . , I.

24. The process for supporting computer infrastructure of claim 19, wherein prior to said reading from said third log:

monitoring a network server for at least one said request to utilize said system resources;

monitoring said network server for at least one response, each response of said at least one response corresponding to a unique request of said at least one said request;

in response to having identified said at least one said request, recording a timestamp T corresponding to the time in which said at least one said request was identified; and

in response to having identified said at least one said response, storing said timestamp T in said third log.