Correlation of stack segment intensity in emergent relationships

• US 10,417,111 B2
• Filed: 05/05/2017
• Issued: 09/17/2019
• Est. Priority Date: 05/09/2016
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Chan, Eric S.
• Primary Examiner(s): Ghaffari, Abu Zar
• Application Number:
  US15/588,521
• Publication Number:
  US 20180074854A1
• Time in Patent Office: 865 Days
• Field of Search: 711132, 711170, 711100, 711131, 717124, 717128, 717131, 717127, 712 E9084, 714758, 714746, 718105
• CPC Class Codes: G06F 11/3037 where the computing system component is a memory, e.g. virtual memory, cache accessing, addressing or allocating within memory systems or architectures G06F12/00; checking stores for correct operation G11C29/00
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3471 Address tracing
• CPC Class Codes: G06F 11/36 Preventing errors by testing or debugging software
• CPC Class Codes: G06F 11/3612 by runtime analysis performance monitoring G06F11/3466
• CPC Class Codes: G06F 12/023 Free address space management
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 3/0629 Configuration or reconfiguration of storage systems
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 9/30058 Conditional branch instructions
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5005 to service a request
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/5016 the resource being the memory
• CPC Class Codes: G06F 9/5022 Mechanisms to release resources
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5044 : considering hardware capabilities
• CPC Class Codes: G06F 9/505 : considering the load
• CPC Class Codes: G06F 9/5055 : considering software capabilities, i.e. software resources associated or available to the machine
• CPC Class Codes: G06F 9/5083 : Techniques for rebalancing the load in a distributed system

Abstract

Methods, systems and computer readable medium are provided for sequentially analyzing a series of thread dump samples to estimate the intensity statistic of newly classified stack segments of stack frames. According to one embodiment, a branch point along one or more linearly connected stack frames of a stack segment can be detected, where the stack segment is associated with one or more thread intensity statistic parameters. Upon detecting the branch point along the one or more linearly connected stack frames of the stack segment, the system can split the stack segment into a plurality of new stack segments that each include a subset of the stack frames, where the plurality of new stack segments are referenced by the stack segment. The system can then initialize the one or more thread intensity statistic parameters for each of the new stack segments.

Claims

1. A computer-implemented method, comprising:

determining one or more thread intensity statistic parameters of a stack segment of a thread executed by a cloud infrastructure system, the stack segment referencing a set of linearly connected stack frames, the one or more thread intensity statistic parameters of the stack segment being determined based at least in part on one or more stack traces received from a memory of the cloud infrastructure system, wherein the one or more thread intensity parameters of the stack segment include a statistical measure of invocations of code blocks referenced by the stack segment;

receiving, from the memory, a stack trace;

upon receiving the stack trace, detecting, based at least in part on the stack trace, a branch point along the set of linearly connected stack frames of the stack segment;

upon detecting the branch point along the set of linearly connected stack frames of the stack segment, splitting the stack segment into a plurality of new stack segments each including a subset of the stack frames, each subset of the stack frames being connected to the branch point, wherein the plurality of new stack segments are referenced by the stack segment;

initializing the one or more thread intensity statistic parameters of the plurality of new stack segments;

determining the one or more thread intensity statistic parameters of the plurality of new stack segments separately based at least in part on one or more subsequent stack traces received from the memory, wherein the one or more thread intensity parameters of the plurality of new stack segments include a statistical measure of invocations of code blocks referenced by the plurality of new stack segments;

outputting the one or more thread intensity statistic parameters to a monitoring system to enable monitoring of performance metrics of the cloud infrastructure system;

determining a relationship between the one or more thread intensity statistic parameters of the stack segments and the one or more thread intensity statistic parameters of the new stack segments; and

detecting one or more service level agreements (SLA) violations within the cloud infrastructure system based on the relationship.

2. The computer-implemented method of claim 1, wherein each of the plurality of new stack segments is a child node and the stack segment is a parent node in a hierarchical relationship.

3. The computer-implemented method of claim 2, wherein splitting the stack segment into the plurality of new stack segments comprises:

creating a first new stack segment of the plurality of new segments that references a first subset of linearly connected stack frames of the set of linearly connected stack frames that precedes the branch point;

creating a second new stack segment of the plurality of new segments that references a second subset of linearly connected stack frames of the set of linearly connected stack frames that follows the branch point; and

modifying the stack segment to reference the first new stack segment as a first child node and to reference the second new stack segment as a second child node; and

wherein the first new stack segment and the second new stack segment reference the stack segment as the parent node.

4. The computer-implemented method of claim 1, wherein determining the one or more thread intensity statistic parameters of the stack segment comprises:

receiving a first of the one or more stack traces;

determining that the first of the one or more stack traces includes each of the set of linearly connected stack frames;

creating the stack segment to reference the set of linearly connected stack frames;

for each of the one or more stack traces other than the first of the one or more stack traces;

receiving the stack trace of the one or more stack traces;

determining that the stack trace of the one or more stack traces includes each of the set of linearly connected stack frames or includes none of the set of linearly connected stack frames; and

determining the one or more thread intensity statistic parameters of the stack segment based at least in part on the stack trace of the one or more stack traces.

5. The computer-implemented method of claim 1, wherein initializing the one or more thread intensity statistic parameters of the plurality of new stack segments comprises copying one or more values of the one or more thread intensity statistic parameters of the stack segment to the one or more thread intensity statistic parameters of the plurality of new stack segments.

6. The computer-implemented method of claim 1, wherein the subset of the stack frames in each of the plurality of new stack segments remain linearly connected.

7. The computer-implemented method of claim 1:

wherein a first stack frame is determined to be linearly connected to a second stack frame based on at least one stack trace being determined to include both the first and second stack frames and no stack trace being determined to include only one of the first and second stack frames; and

wherein a third stack frame and a fourth stack frame are determined to be not linearly connected based on another at least one stack trace being determined to include only the one of the third stack frame or the fourth stack frame.

8. The computer-implemented method of claim 1, wherein the one or more thread intensity statistic parameters include:

a number of occurrences of the set of linearly connected stack frames;

one or more measures of intensity of the set of linearly connected stack frames, wherein each of the one or more measures of intensity is associated with a different season of a time period;

one or more seasonal factors, wherein each of the one or more seasonal factors is associated with one of the one or more measures of intensity of the set of linearly connected stack frames;

or a combination thereof.

9. The computer-implemented method of claim 1, wherein the one or more stack traces, the stack trace, and the one or more subsequent stack traces are obtained via one or more thread dumps taken during an execution of a multi-threaded program.

10. The computer-implemented method of claim 1,

wherein the one or more SLA violations are detected based on detecting performance glitches within the cloud infrastructure system based on the relationship.

11. A system comprising:

one or more processors; and

a memory accessible to the one or more processors, the memory storing one or more instructions that, upon execution by the one or more processors, causes the one or more processors to;

determine one or more thread intensity statistic parameters of a stack segment of a thread executed by a cloud infrastructure system, the stack segment referencing a set of linearly connected stack frames, the one or more thread intensity statistic parameters of the stack segment being determined based at least in part on one or more stack traces received from the memory, wherein the one or more thread intensity parameters of the stack segment include a statistical measure of invocations of code blocks referenced by the stack segment;

receive, from the memory, a stack trace;

upon receiving the stack trace, detect, based at least in part on the stack trace, a branch point along the set of linearly connected stack frames of the stack segment;

upon detecting the branch point along the set of linearly connected stack frames of the stack segment, split the stack segment into a plurality of new stack segments each including a subset of the stack frames, each subset of the stack frames being connected to the branch point, wherein the plurality of new stack segments are referenced by the stack segment;

initialize the one or more thread intensity statistic parameters of the plurality of new stack segments;

determine the one or more thread intensity statistic parameters of the plurality of new stack segments separately based at least in part on one or more subsequent stack traces received from the memory, wherein the one or more thread intensity parameters of the plurality of new stack segments include a statistical measure of invocations of code blocks referenced by the plurality of new stack segments;

output the one or more thread intensity statistic parameters to a monitoring system to enable monitoring of performance metrics of the cloud infrastructure system;

determine a relationship between the one or more thread intensity statistic parameters of the stack segments and the one or more thread intensity statistic parameters of the new stack segments; and

detect one or more service level agreements (SLA) violations within the cloud infrastructure system based on the relationship.

12. The system of claim 11, wherein each of the plurality of new stack segments is a child node and the stack segment is a parent node in a hierarchical relationship.

13. The system of claim 12, wherein splitting the stack segment into the plurality of new stack segments comprises:

creating a first new stack segment of the plurality of new segments that references a first subset of linearly connected stack frames of the set of linearly connected stack frames that precedes the branch point;

creating a second new stack segment of the plurality of new segments that references a second subset of linearly connected stack frames of the set of linearly connected stack frames that follows the branch point; and

modifying the stack segment to reference the first new stack segment as a first child node and to reference the second new stack segment as a second child node; and

wherein the first new stack segment and the second new stack segment reference the stack segment as the parent node.

14. The system of claim 11, wherein determining the one or more thread intensity statistic parameters of the stack segment comprises:

receiving a first of the one or more stack traces;

determining that the first of the one or more stack traces includes each of the set of linearly connected stack frames;

creating the stack segment to reference the set of linearly connected stack frames;

for each of the one or more stack traces other than the first of the one or more stack traces;

receiving the stack trace of the one or more stack traces;

determining that the stack trace of the one or more stack traces includes each of the set of linearly connected stack frames or includes none of the set of linearly connected stack frames; and

determining the one or more thread intensity statistic parameters of the stack segment based at least in part on the stack trace of the one or more stack traces.

15. The system of claim 11, wherein initializing the one or more thread intensity statistic parameters of the plurality of new stack segments comprises copying one or more values of the one or more thread intensity statistic parameters of the stack segment to the one or more thread intensity statistic parameters of the plurality of new stack segments.

16. The system of claim 11, wherein the subset of the stack frames in each of the plurality of new stack segments remain linearly connected.

17. The system of claim 11:

wherein a first stack frame is determined to be linearly connected to a second stack frame based on at least one stack trace being determined to include both the first and second stack frames and no stack trace being determined to include only one of the first and second stack frames; and

wherein a third stack frame and a fourth stack frame are determined to be not linearly connected based on another at least one stack trace being determined to include only the one of the third stack frame or the fourth stack frame.

18. The system of claim 11, wherein the one or more thread intensity statistic parameters include:

a number of occurrences of the set of linearly connected stack frames;

one or more measures of intensity of the set of linearly connected stack frames, wherein each of the one or more measures of intensity is associated with a different season of a time period;

one or more seasonal factors, wherein each of the one or more seasonal factors is associated with one of the one or more measures of intensity of the set of linearly connected stack frames;

or a combination thereof.

19. The system of claim 11, wherein the one or more stack traces, the stack trace, and the one or more subsequent stack traces are obtained via one or more thread dumps taken during an execution of a multi-threaded program.

20. A non-transitory computer-readable medium storing one or more instructions that, upon execution by one or more processors, causes the one or more processors to:

determine one or more thread intensity statistic parameters of a stack segment of a thread executed by a cloud infrastructure system, the stack segment referencing a set of linearly connected stack frames, the one or more thread intensity statistic parameters of the stack segment being determined based at least in part on one or more stack traces received from a memory of the cloud infrastructure system, wherein the one or more thread intensity parameters of the stack segment include a statistical measure of invocations of code blocks referenced by the stack segment;

receive, from the memory, a stack trace;

upon receiving the stack trace, detect, based at least in part on the stack trace, a branch point along the set of linearly connected stack frames of the stack segment;

upon detecting the branch point along the set of linearly connected stack frames of the stack segment, split the stack segment into a plurality of new stack segments each including a subset of the stack frames, each subset of the stack frames being connected to the branch point, wherein the plurality of new stack segments are referenced by the stack segment;

initialize the one or more thread intensity statistic parameters of the plurality of new stack segments;

determine the one or more thread intensity statistic parameters of the plurality of new stack segments separately based at least in part on one or more subsequent stack traces received from the memory, wherein the one or more thread intensity parameters of the plurality of new stack segments include a statistical measure of invocations of code blocks referenced by the plurality of new stack segments;

output the one or more thread intensity statistic parameters to a monitoring system to enable monitoring of performance metrics of the cloud infrastructure system;

determine a relationship between the one or more thread intensity statistic parameters of the stack segments and the one or more thread intensity statistic parameters of the new stack segments; and

detect one or more service level agreements (SLA) violations within the cloud infrastructure system based on the relationship.