System and method for memory leak detection in a virtual machine environment

• US 7,757,202 B2
• Filed: 11/05/2007
• Issued: 07/13/2010
• Est. Priority Date: 12/20/2002
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: BEA Systems Incorporated (Oracle Corporation)
• Inventors: Vidstedt, Mikael, Lonnebring, Peter, Dahlstedt, Joakim
• Primary Examiner(s): Wang; Philip
• Application Number:
  US11/935,287
• Publication Number:
  US 20080065842A1
• Time in Patent Office: 981 Days
• Field of Search: 717/100, 707/206
• US Class Current: 717/100
• CPC Class Codes: G06F 12/0253 Garbage collection, i.e. reclamation of unreferenced memory
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99957 Garbage collection

Abstract

A system and method for detecting memory leaks in a garbage collected environment. In particular, the present invention provides a mechanism by which objects in the memory space of a virtual machine are analyzed as to whether they are warm or cold, i.e., how long they have persisted in memory without being accessed or referenced by another object. Following this analysis, the objects are clustered into groups of warm or cold objects, and any links from warm clusters to cold clusters are determined. This information is then provided to the software developer for use in detecting and analyzing potential memory leaks.

Claims

1. A method for detecting potential memory leaks in a run-time environment, the run-time environment including a virtual machine and a memory for storing objects, the method comprising the steps of:

accepting as input from a system developer a value for a limiting time;

determining, for each of a plurality of objects in memory, a length of time since that object has persisted in memory without being accessed or referenced, and

if the object has persisted in memory without being accessed or referenced in memory greater than the limiting time, then identifying the object as one of a first object type,

if the object has persisted in memory without being accessed or referenced in memory for less than the limiting time, then identifying the object as one of a second object type; and

determining links from any object of the first object type to any object of the second object type, and displaying the links for use by the system developer in detecting potential memory leaks.

2. The method of claim 1, further comprising the step of:

clustering groups of the first type of objects to form first clusters, and groups of the second type of objects to form second clusters.

3. The method of claim 2, further comprising the step of:

reducing the number of first and second clusters to those that occupy more than a particular amount of memory space.

4. The method of claim 2, wherein the links include any or both of object of the first type—

object of the second type links and first cluster—

second cluster links.

5. The method of claim 1, further comprising the step of:

reporting information about the links for use by the system developer in determining potential memory leaks.

6. The method of claim 1, wherein displaying the links comprises displaying the links on a computer screen device.

7. The method of claim 2, further comprising the step of:

adjusting by the system developer the limiting time in determining whether an object is of a first type or of a second type to better distinguish between first and second types of objects or first and second clusters.

8. The method of claim 2, wherein the objects are not moved in the memory when clustered.

9. A system for detecting potential memory leaks in a run-time environment, the system comprising:

a virtual machine executing within the run-time environment;

a memory within the run-time environment for storing objects, for use by a software application; and

an analyzer that accepts as input from a system developer a value for a limiting time;

wherein the analyzer determines, for each of a plurality of objects in memory, a length of time since that object has persisted in memory without being accessed or referenced, and

if the object has persisted in memory without being accessed or referenced in memory greater than the limiting time, then the object is identified as one of a first object type,

if the object has persisted in memory without being accessed or referenced in memory for less than the limiting time, then the object is identified as one of a second object type; and

wherein the analyzer determines links from any object of the first object type to any object of the second object type, and the links are displayed for use by the system developer in detecting potential memory leaks.

10. The system of claim 9 further comprising:

an object clusterer for clustering groups of the first type of objects to form first clusters, and groups of the second type of objects to form second clusters.

11. The system of claim 10, wherein the number of first and second clusters are reduced to those that occupy more than a particular amount of memory space.

12. The system of claim 10, wherein the links include any or both of object of the first type—

object of the second type links and first cluster—

second cluster links.

13. The system of claim 9, wherein information about the links is reported to the system developer for use in determining potential memory leaks.

14. The system of claim 9, wherein the links are displayed on a computer screen device.

15. The system of claim 10, wherein the limiting time determining whether an object is of a first type or of a second type is adjusted by the system developer to better distinguish between first and second types of objects or first and second clusters.

16. The system of claim 10, wherein the objects are not moved in the memory when clustered.

17. A computer readable medium, including instructions stored thereon, which when executed by a computer cause the computer to perform steps for method for detecting potential memory leaks in a run-time environment, said run-time environment including a virtual machine and a memory for storing objects, the steps comprising:

accepting as input from a system developer a value for a limiting time;

determining, for each of a plurality of objects in memory, a length of time since that object has persisted in memory without being accessed or referenced, and

if the object has persisted in memory without being accessed or referenced in memory greater than the limiting time, then identifying the object as one of a first object type,

if the object has persisted in memory without being accessed or referenced in memory for less than the limiting time, then identifying the object as one of a second object type; and

determining links from any object of the first object type to any object of the second object type, and displaying the links for use by the system developer in detecting potential memory leaks.

18. The computer readable medium of claim 17, further comprising the step of:

clustering groups of the first type of objects to form first clusters, and groups of the second type of objects to form second clusters.

19. The computer readable medium of claim 18, further comprising the step of:

reducing the number of first and second clusters to those that occupy more than a particular amount of memory space.

20. The computer readable medium of claim 18, wherein the links include any or both of object of the first type—

object of the second type links and first cluster—

second cluster links.

21. The computer readable medium of claim 17, further comprising the step of:

reporting information about the links for use by the system developer in determining potential memory leaks.

22. The computer readable medium of claim 17, wherein displaying the links comprises displaying the links on a computer screen device.

23. The computer readable medium of claim 18, further comprising the step of:

adjusting by the system developer the limiting time in determining whether an object is of a first type or of a second type to better distinguish between first and second types of objects or first and second clusters.

24. The computer readable medium of claim 18, wherein the objects are not moved in the memory when clustered.