Memory Leak Detection Using Transient Workload Detection And Clustering

A system, comprising:

• one or more computing devices configured to:
• classify, based at least in part on a first statistical analysis, one or more performance metric collections corresponding to an observation period of an executing process into a first set of metrics representing one or more sub-periods during which a workload of the executing process exhibited a change trend, and a second set of metrics representing one or more sub-periods during which the workload did not exhibit a change trend;
• group, based at least in part on a second statistical analysis, the second set of metrics into one or more metric clusters, wherein each metric cluster of the one or more metric clusters represents a respective statistically stable workload level;
• determine, based at least in part on a third statistical analysis of a particular performance metric corresponding to the one or more metric clusters, whether an upward trend in memory usage at the executing process during the observation time period is indicated;
• and
• in response to a determination that an upward trend in memory usage is indicated, provide a notification of a potential memory leak at the executing process.

A non-transitory computer-accessible storage medium storing program instructions that when executed on one or more processors:

• classify, based at least in part on a first statistical analysis, one or more performance metric collections corresponding to an observation period of an executing process into a first set of metrics representing one or more sub-periods during which a workload of the executing process exhibited a change trend, and a second set of metrics representing one or more sub-periods during which the workload did not exhibit a change trend;
• determine, based at least in part on a second statistical analysis of the second set of metrics, one or more metric clusters, wherein each metric cluster of the one or more metric clusters represents a respective statistically stable workload level;
• detect, based at least in part on a third statistical analysis of a particular performance metric corresponding to the one or more metric clusters, whether an upward trend in memory usage at the executing process during the observation time period is indicated;
• and
• in response to a detection that an upward trend in memory usage at the executing process during the observation time period is indicated, provide a notification of a potential memory leak at the executing process.

A method, comprising:

• performing, by one or more computing devices:
• collecting a set of metrics of an executing process over an observation period;
• identifying, using at least a particular metric of the set, one or more sub-periods of the observation period corresponding to a statistically stable workload level;
• determining, using at least a different metric of the set, whether memory usage at the executing process increased during the one or more sub-periods corresponding to the statistically stable workload level;
• and
• in response to determining that the memory usage at the executing process increased during the one or more sub-periods, generating a notification of a potential memory leak at the executing process.

A non-transitory computer-accessible storage medium storing program instructions that when executed on one or more processors:

• collect a set of metrics of an executing process over an observation period;
• identify, using at least a particular metric of the set, one or more sub-periods of the observation period corresponding to a statistically stable workload level;
• determine, using at least a different metric of the set, whether memory usage at the executing process increased during the one or more sub-periods corresponding to the statistically stable workload level;
• and
• in response to a determination that the memory usage at the executing process increased during the one or more sub-periods, generate a notification of a potential memory leak at the executing process.