System and method for monitoring performance of a server system using otherwise unused processing capacity of user computing devices

• US 6,738,813 B1
• Filed: 09/11/2000
• Issued: 05/18/2004
• Est. Priority Date: 09/11/2000
• Status: Expired due to Term

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Mercury Interactive Corp. (HP Inc.)
• Inventors: Reichman, David
• Primary Examiner(s): Winder, Patrice
• Assistant Examiner(s): Tran, Philip B.
• Application Number:
  US09/659,476
• Time in Patent Office: 1,345 Days
• Field of Search: 709/224, 709/223, 709/225, 709/238, 709/241, 709/105
• US Class Current: 709/224
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: H04L 41/046 comprising network management agents or mobile agents therefor
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 43/12 Network monitoring probes

Abstract

A monitoring system provides a service for users to monitor their respective Web sites, or other server systems, as seen from the computing devices of other users. In a preferred embodiment, the system includes an agent component that runs on the computing devices of service users to provide functionality for accessing and monitoring the performance of a server (preferably when such devices are otherwise idle). By running the agent component on a computer, a user effectively makes that computer available to others for use as a monitoring agent. Processing resources for conducting monitoring sessions are thus contributed and shared by members of a community. The agents are remotely configurable over the Internet, and may be configured, for example, to execute a particular Web transaction while monitoring specified performance parameters (server response times, network hop delays, server availability, etc). Using a service provider Web site, a user of the service can set up a monitoring session in which agent devices of other community members are used to monitor the performance of the user'"'"'s server system. In one embodiment, the system implements a reciprocity policy in which the extent to which a user may use the service for monitoring is dependent upon the quantity of processing resources (e.g., transaction execution events) that user has contributed to the community. Performance data collected by the agents may be used both to generate server-specific reports and general Internet “weather maps.”

Claims

1. A method of monitoring performance of a server system, the method comprising:

monitoring CPU utilization of each of a plurality of user computers that are distributed geographically on a computer network to identify a group of said user computers that are in a substantially idle state, said group comprising multiple user computers;

causing the substantially idle user computers in the group to access a target server system over the computer network as simulated users thereof while monitoring performance of the target server system, such that each of the user computers in the group generates performance data reflective of the performance of the target server system as seen from the respective locations of such user computers; and

collecting the performance data from each of the user computers in the group;

whereby otherwise unused processing capacity of the user computers is used to monitor the target server system.

2. The method of claim 1, wherein the group of user computers comprises computers in multiple different geographic locations, and the user computers monitor end user performance of the target server system as seen from each of the multiple geographic locations.

3. The method of claim 1, wherein causing the substantially idle user computers in the group to access the target server system comprises sending work requests over the computer network to the user computers in the group in response to detecting that each such user computer is in a substantially idle state.

4. The method of claim 1, wherein causing the substantially idle user computers in the group to access the target server system comprises executing, on each user computer, an agent component that initiates an access to the target server system when its respective user computer is substantially idle.

5. The method of claim 1, wherein the CPU utilization of each of the plurality of user computers is monitored using an agent component installed on each of the plurality of user computers, said agent component including functionality for communicating with target server systems as a simulated user.

6. The method of claim 5, wherein the agent component runs as a background task on each of the plurality of user computers.

7. The method of claim 1, wherein the method comprises controlling the user computers in the group so as to apply a controlled load to the target server system, to thereby load test the target server system.

8. The method of claim 1, further comprising determining whether a quantity of processing resources provided by the group of user computers is sufficient to apply a specified load to the target server system.

9. The method of claim 1, wherein the performance data collected from the user computers in the group includes hop delay data for network hops between the target server system and specific user computers.

10. The method of claim 9, further comprising aggregating the hop delay data received from multiple user computers for a common hop, and using the aggregated hop delay data to evaluate performance of a router associated with the hop.

11. The method of claim 1, further comprising maintaining, for at least one user computer in the group, a record reflective of a quantity of that user computer'"'"'s resources used to monitor the target server system.

12. The method of claim 11, further comprising using the record to control an extent to which an associated user can use the plurality of user computers to monitor a second target server system.

13. A server monitoring system that operates according to the method of claim 1.

14. A server monitoring system, comprising:

an agent component that is adapted to run on user computing devices to form agents for monitoring target server systems, wherein the agent component includes functionality for monitoring processing loads of the user computing devices on which the agent component runs, and further includes functionality for monitoring server performance by at least (a) sending request messages to specific target server systems as simulated users thereof, and (b) monitoring responses to such request messages to generate performance data, wherein the agent component is configured to monitor end user performance only when a host computing device on which the agent component runs is in a substantially idle state;

a controller that sends work requests to the agents over a computer network, the work requests specifying request messages to be sent by the agents to specific target server systems; and

a reports server that generates performance reports based on the performance data generated by the agents;

wherein the agent component takes processing loads of the user computing devices into consideration in determining when to send the requests messages to the target server system such that the target server system is monitored primarily using otherwise-unused processing resources of the user computing devices.

15. The server monitoring system of claim 14, wherein the agent component runs as a background task on the user computing devices.

16. The server monitoring system of claim 14, wherein the agent component monitors performance of a target server system from a host computing device only when a processing load of the host computing device falls below a specified threshold, so that the agent component does not substantially interfere with ordinary operation of the host computing device.

17. The server monitoring system of claim 14, wherein the controller controls the agents so as to apply a controlled load to a target server system, to thereby load-test the target server system.

18. The system as in claim 14, wherein the agents measure and report network hop delays, and the reports server aggregates hop delay data collected by the agents over multiple monitoring sessions to generate a general network status report.

19. The server monitoring system of claim 14, wherein the controller takes into consideration the geographic locations of the agents in selecting a set of agents to use to monitor a target server system.

20. The server monitoring system of claim 14, wherein the controller takes into consideration the current processing loads of the agents in selecting a set of agents to use to monitor a target server system.

21. The server monitoring system of claim 14, wherein the controller supports an ability to set up a monitoring session in which a plurality of geographically distributed agents are used to access and monitor a target server system, such that end user performance of the target server system may be monitored as seen from multiple geographic locations.

22. The system as in claim 14, wherein the controller implements a reciprocity policy in which a user of the system is given credit toward use of the system when a computing device of that user is used as an agent within a monitoring session of another user.

23. The system as in claim 14, further comprising a user interface that displays a directory of agents that are available for use, and provides user functionality for manually selecting agents to include in a monitoring session.