Back to results
US 10,708,344 B1Granted
Continuous performance management in information processing systems
AssigneeEMC IP Holding Company LLC
InventorsJunping Zhao, Yu Cao, Sanping Li, Kenneth Durazzo
Life of this patent
Estimating the term…
Abstract
At least one performance metric is monitored in an information processing system to detect a performance event substantially contemporaneous with the occurrence of the event, wherein monitoring of the performance metric is performed in a continuous manner. At least a portion of the information processing system is profiled in response to detection of a performance event, wherein the profiling step obtains a performance snapshot of the portion of the information processing system being profiled. The performance snapshot is analyzed to determine a root cause of the performance event within the information processing system.
Claims20
A method comprising:
- monitoring at least one performance metric in an information processing system to detect a performance event substantially contemporaneous with the occurrence of the event, wherein the monitoring of the at least one performance metric is performed in a continuous manner;
- wherein the performance event is detected based at least in part on whether a current value of the at least one performance metric deviates from a baseline for the at least one performance metric and an amount that the current value deviates from the baseline, and wherein the current value is determined based on an aggregation of values collected for the at least one performance metric over a given time interval;
- profiling at least a portion of the information processing system in response to the detection of the performance event, wherein the profiling step comprises obtaining a performance snapshot of the portion of the information processing system being profiled;
- analyzing the performance snapshot to determine a root cause of the performance event within the information processing system;
- and
- determining a load pressure of the information processing system and, based on the determined load pressure, configuring a sensitivity of a sampling granularity of the information processing system;
- wherein the performance event is represented by a tuple comprising a plurality of elements comprising a first element representing the aggregation of values for the at least one performance metric over the given time interval, a second element representing whether the current value is better or worse than the baseline, a third element representing the percentage that the current value deviates from the baseline and a contemporaneous performance statistic with a configurable weight, and a fourth element representing a time interval in which to confirm the amount that the current value deviates from the baseline;
- wherein the profiling step is incremental as determined by a severity of the performance event;
- wherein the profiling step further comprises profiling one or more components of the information processing system based on the performance event;
- wherein a number of the one or more components of the information processing system being profiled incrementally increases as the severity of the performance event increases;
- wherein a sensitivity of the sampling granularity increases to a fine-granularity to detect ephemeral performance fluctuations in response to a determination of a decrease in the load pressure;
- wherein, following the increase to the fine-granularity, the sensitivity of the sampling granularity decreases to a coarse-granularity in response to a determination of an increase in the load pressure;
- and
- wherein the monitoring, profiling, analyzing, and determining steps are performed by a processing device comprising a processor operatively coupled to a memory.
11 more elements
A system comprising:
- a memory;
- and
- a processor operatively coupled to the memory and configured to:
- monitor at least one performance metric in an information processing system to detect a performance event substantially contemporaneous with the occurrence of the event, wherein the monitoring of the at least one performance metric is performed in a continuous manner;
- wherein the performance event is detected based at least in part on whether a current value of the at least one performance metric deviates from a baseline for the at least one performance metric and an amount that the current value deviates from the baseline, and wherein the current value is determined based on an aggregation of values collected for the at least one performance metric over a given time interval;
- profile at least a portion of the information processing system in response to the detection of the performance event, wherein, in performing the profiling, the processor is configured to obtain a performance snapshot of the portion of the information processing system being profiled;
- analyze the performance snapshot to determine a root cause of the performance event within the information processing system;
- and
- determine a load pressure of the information processing system and, based on the determined load pressure, configure a sensitivity of a sampling granularity of the information processing system;
- wherein the performance event is represented by a tuple comprising a plurality of elements comprising a first element representing the aggregation of values for the at least one performance metric over the given time interval, a second element representing whether the current value is better or worse than the baseline, a third element representing the percentage that the current value deviates from the baseline and a contemporaneous performance statistic with a configurable weight, and a fourth element representing a time interval in which to confirm the amount that the current value deviates from the baseline;
- wherein the profiling is incremental as determined by a severity of the performance event;
- wherein, in performing the profiling, the processor is further configured to profile one or more components of the information processing system based on the performance event;
- wherein a number of the one or more components of the information processing system being profiled incrementally increases as the severity of the performance event increases;
- wherein a sensitivity of the sampling granularity increases to a fine-granularity to detect ephemeral performance fluctuations in response to a determination of a decrease in the load pressure;
- and
- wherein, following the increase to the fine-granularity, the sensitivity of the sampling granularity decreases to a coarse-granularity in response to a determination of an increase in the load pressure.
13 more elements
An article of manufacture comprising a processor-readable non-transitory storage medium having encoded therein executable code of one or more software programs, wherein the one or more software programs when executed by one or more processing devices implement steps of:
- monitoring at least one performance metric in an information processing system to detect a performance event substantially contemporaneous with the occurrence of the event, wherein the monitoring of the at least one performance metric is performed in a continuous manner;
- wherein the performance event is detected based at least in part on whether a current value of the at least one performance metric deviates from a baseline for the at least one performance metric and an amount that the current value deviates from the baseline, and wherein the current value is determined based on an aggregation of values collected for the at least one performance metric over a given time interval;
- profiling at least a portion of the information processing system in response to the detection of the performance event, wherein the profiling step comprises obtaining a performance snapshot of the portion of the information processing system being profiled;
- analyzing the performance snapshot to determine a root cause of the performance event within the information processing system;
- and
- determining a load pressure of the information processing system and, based on the determined load pressure, configuring a sensitivity of a sampling granularity of the information processing system;
- wherein the performance event is represented by a tuple comprising a plurality of elements comprising a first element representing the aggregation of values for the at least one performance metric over the given time interval, a second element representing whether the current value is better or worse than the baseline, a third element representing the percentage that the current value deviates from the baseline and a contemporaneous performance statistic with a configurable weight, and a fourth element representing a time interval in which to confirm the amount that the current value deviates from the baseline;
- wherein the profiling step is incremental as determined by a severity of the performance event;
- wherein the profiling step further comprises profiling one or more components of the information processing system based on the performance event;
- wherein a number of the one or more components of the information processing system being profiled incrementally increases as the severity of the performance event increases;
- wherein a sensitivity of the sampling granularity increases to a fine-granularity to detect ephemeral performance fluctuations in response to a determination of a decrease in the load pressure;
- and
- wherein, following the increase to the fine-granularity, the sensitivity of the sampling granularity decreases to a coarse-granularity in response to a determination of an increase in the load pressure.
10 more elements
Dependencies are inferred from the claim text.