Adaptive performance data measurement and collections

• US 20020183972A1
• Filed: 06/01/2001
• Published: 12/05/2002
• Est. Priority Date: 06/01/2001
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Enck, Brent A., Knoop, James A., Flood, David S., Martin, Andrea, Green, John
• Granted Patent: US 6,609,083 B2
• Time in Patent Office: Days
• US Class Current: 702/186
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3072 where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/348 Circuit details, i.e. tracer hardware
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: H04L 41/5019 Ensuring fulfilment of SLA
• CPC Class Codes: H04L 43/55 Testing of service level quality, e.g. simulating service usage

Abstract

Methods and systems that adaptively measure and collect performance data are disclosed. An exemplary method includes acquiring system performance data in accordance with a performance policy. Then, data measurement and collection is dynamically adjusted based on the performance policy and a detected system condition. An alternative method includes establishing a self-monitoring data collection policy. At least one system performance metric is then measured in accordance with the self-monitoring data collection policy. The at least one system performance metric is evaluated to determine system health. The system performance metric measuring step is then dynamically adjusted based on the self-monitoring data collection policy and the system health. The steps of measuring, evaluating and dynamically adjusting are then repeated.

Claims

1. A method of adaptively measuring performance data in a computer-based system comprising:

acquiring system performance data in accordance with a performance policy; and

dynamically adjusting data measurement and collection based on the performance policy and a detected system condition.

2. The method of claim 1, wherein adjusting data measurement and collection comprises at least one of adjusting a frequency of data collection and adjusting data quantity collected at each data collection interval.

3. The method of claim 2, further comprising:

resetting at least one of the frequency of data collection, and the data quantity collected at each data collection interval to a base level when the detected system condition does not violate the performance policy.

4. The method of claim 2, further comprising:

increasing at least one of the frequency of data collection and the data quantity collected at each data collection interval when the detected system condition violates the performance policy.

5. The method of claim 4, wherein the detected system condition is at least one of a CPU bottleneck, a memory bottleneck and a disk bottleneck.

6. The method of claim 5, wherein the performance policy is violated if a threshold is exceeded, wherein the threshold is at least one of a percentage of total processor time, a percentage of committed memory in use, and a percentage of disk use time.

7. The method of claim 1, further comprising:

storing data in at least one of a persistent storage element and a transient storage element.

8. The method of claim 1, wherein acquiring system performance data is performed in at least one of a fixed time manner, a relative time manner and an event based manner.

9. The method of claim 1, further comprising:

collecting data in at least one of a time series collection, a baseline collection and an event based collection.

10. A method of adaptively monitoring performance data in a computer-based system comprising:

establishing a self-monitoring data collection policy;

measuring at least one system performance metric in accordance with the self-monitoring data collection policy;

evaluating the at least one system performance metric to determine system health;

dynamically adjusting the system performance metric measuring step based on the self-monitoring data collection policy and the system health; and

repeating the steps of measuring, evaluating and dynamically adjusting.

11. An adaptative measurement agent for measuring performance data in a computer-based system comprising:

logic that acquires system performance data in accordance with a performance policy; and

logic that dynamically adjusts data measurement and collection based on the performance policy and a detected system condition.

12. The adaptative measurement agent of claim 11, wherein adjusting data measurement and collection comprises at least one of adjusting a frequency of data collection and adjusting data quantity collected at each data collection interval.

13. The adaptative measurement agent of claim 12, further comprising:

logic that resets at least one of the frequency of data collection, and the data quantity collected at each data collection interval to a base level when the detected system condition does not violate the performance policy.

14. The adaptative measurement agent of claim 12, further comprising:

logic that increases at least one of the frequency of data collection and the data quantity collected at each data collection interval when the detected system condition violates the performance policy.

15. The adaptative measurement agent of claim 14, where in the detected system condition i s at least one of a CPU bottleneck, a memory bottleneck and a disk bottleneck.

16. The adaptative measurement agent of claim 15, wherein the performance policy is violated if a threshold is exceeded, wherein the threshold is at least one of a percentage of total processor time, a percentage of committed memory in use, and a percentage of disk use time.

17. The adaptative measurement agent of claim 11, further comprising:

logic that stores data in at least one of a persistent storage element and a transient storage element.

18. The adaptative measurement agent of claim 11, wherein acquiring system performance data is performed in at least one of a fixed time manner, a relative time manner and an event based manner.

19. The adaptative measurement agent of claim 11, further comprising:

logic that collects data in at least one of a time series collection, a baseline collection and an event based collection.

20. The adaptative measurement agent of claim 11, wherein the measurement agent is integrated in a network management system.