Method and apparatus for computer network analysis

• US 20020147937A1
• Filed: 08/01/2001
• Published: 10/10/2002
• Est. Priority Date: 01/16/2001
• Status: Active Grant

• Current Assignee: Infonet Services Corp. (BT Group PLC)
• Original Assignee: Infonet Services Corp. (BT Group PLC)
• Inventors: Wolf, Gregory H.
• Granted Patent: US 7,058,843 B2
• Time in Patent Office: Days
• US Class Current: 714/4
• CPC Class Codes: H04L 41/0806 for initial configuration or provisioning, e.g. plug-and-play
• CPC Class Codes: H04L 41/22 comprising specially adapted graphical user interfaces [GUI]
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• CPC Class Codes: H04L 41/5096 wherein the managed service relates to distributed or central networked applications
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/0882 Utilisation of link capacity
• CPC Class Codes: H04L 43/50 Testing arrangements

Abstract

A method and system for computer network analysis. A test network is configured and test scenarios are run using exemplary software applications. Trace data including network response and load information is collected during the test scenarios. The trace data is used as input data into various network analysis tools. The network analysis tools evaluate the computer network'"'"'s response using a variety of criteria. A final report is generated proposing a computer network configuration suitable for satisfying the requirements of the exemplary software applications.

Claims

1. A method for determining a computer network'"'"'s performance during operation of a software application using the computer network, comprising:

recording network traffic data while the software application is using the computer network;

generating from the network traffic data a latency sensitivity metric;

generating from the network traffic data a bandwidth sensitivity metric; and

comparing the latency sensitivity metric and the bandwidth sensitivity metric to determine the computer network'"'"'s performance during operation of the software application.

2. The method of claim 1, wherein generating from the network traffic data the latency sensitivity metric includes:

generating from the network traffic data a plurality of computer network response times for a plurality of software application use scenarios at a constant computer network bandwidth value; and

generating the latency sensitivity metric from the plurality of computer network response times.

3. The method of claim 2, wherein the latency sensitivity metric is generated by calculating the standard deviation of the plurality of network response times.

4. The method of claim 2, wherein the latency sensitivity metric is generated by calculating the slope of a line defined by plotting the plurality of computer network response times versus the plurality of software application use scenarios.

5. The method of claim 1, wherein generating from the network traffic data the bandwidth sensitivity metric includes:

generating from the network traffic data a plurality of computer network response times for a plurality of computer network bandwidth values for a software application use scenario; and

generating the bandwidth sensitivity metric from the plurality of computer network response times.

6. The method of claim 5, wherein the bandwidth sensitivity metric is generated by calculating the standard deviation of the plurality of network response times.

7. The method of claim 5, wherein the bandwidth sensitivity metric is generated by calculating the slope of a line defined by plotting the plurality of computer network response times versus the plurality of computer network bandwidth values.

8. The method of claim 1, the method further comprising generating from the network traffic data a return on investment data table for use in generating a return on investment metric.

9. The method of claim 1, the method further comprising generating from the network traffic data a computer network simulation table for use in simulating a second computer network.

10. A data processing system adapted to determine a computer network'"'"'s performance during operation of a software application using the computer network, comprising:

a processor; and

a memory operably coupled to the processor and having program instructions stored therein, the processor being operable to execute the program instructions, the program instructions including;

receiving network traffic data recorded while the software application is using the computer network;

generating from the network traffic data a latency sensitivity metric;

generating from the network traffic data a bandwidth sensitivity metric; and

comparing the latency sensitivity metric and the bandwidth sensitivity metric to determine the computer network'"'"'s performance during operation of the software application.

11. The data processing system of claim 10, wherein the program instructions for generating from the network traffic data the latency sensitivity metric further include:

generating from the network traffic data a plurality of computer network response times for a plurality of software application use scenarios at a constant computer network bandwidth value; and

generating the latency sensitivity metric from the plurality of computer network response times.

12. The data processing system of claim 11, wherein the latency sensitivity metric is generated by calculating the standard deviation of the plurality of network response times.

13. The data processing system of claim 11, wherein the latency sensitivity metric is generated by calculating the slope of a line defined by plotting the plurality of computer network response times versus the plurality of software application use scenarios.

14. The data processing system of claim 10, wherein the program instructions for generating from the network traffic data the bandwidth sensitivity metric further include:

generating from the network traffic data a plurality of computer network response times for a plurality of computer network bandwidth values for a software application use scenario; and

generating the bandwidth sensitivity metric from the plurality of computer network response times.

15. The data processing system of claim 14, wherein the bandwidth sensitivity metric is generated by calculating the standard deviation of the plurality of network response times.

16. The data processing system of claim 14, wherein the bandwidth sensitivity metric is generated by calculating the slope of a line defined by plotting the plurality of computer network response times versus the plurality of computer network bandwidth values.

17. The data processing system of claim 10, the program instructions further including generating from the network traffic data a return on investment data table for use in generating a return on investment metric.

18. The data processing system of claim 10, the program instructions further including generating from the network traffic data a computer network simulation table for use in simulating a second computer network.

20. The computer readable media of claim 19, wherein the program instructions for generating from the network traffic data the latency sensitivity metric further include:

generating from the network traffic data a plurality of computer network response times for a plurality of software application use scenarios at a constant computer network bandwidth value; and

generating the latency sensitivity metric from the plurality of computer network response times.

21. The computer readable media of claim 20, wherein the latency sensitivity metric is generated by calculating the standard deviation of the plurality of network response times.

22. The computer readable media of claim 21, wherein the latency sensitivity metric is generated by calculating the slope of a line defined by plotting the plurality of computer network response times versus the plurality of software application use scenarios.

23. The computer readable media of claim 19, wherein the program instructions for generating from the network traffic data the bandwidth sensitivity metric further include:

generating from the network traffic data a plurality of computer network response times for a plurality of computer network bandwidth values for a software application use scenario; and

generating the bandwidth sensitivity metric from the plurality of computer network response times.

24. The computer readable media of claim 23, wherein the bandwidth sensitivity metric is generated by calculating the standard deviation of the plurality of network response times.

25. The computer readable media of claim 23, wherein the bandwidth sensitivity metric is generated by calculating the slope of a line defined by plotting the plurality of computer network response times versus the plurality of computer network bandwidth values.

26. The computer readable media of claim 19, the program instructions further comprising generating from the network traffic data a return on investment data table for use in generating a return on investment metric.

27. The computer readable media of claim 19, the program instructions further comprising generating from the network traffic data a computer network simulation table for use in simulating a second computer network.

19. A computer readable media embodying computer program instructions for execution by a computer, the computer program instructions adapting a computer to determine a computer network'"'"'s performance during operation of a software application using the computer network, the computer instructions comprising:

receiving network traffic data recorded while the software application is using the computer network;

generating from the network traffic data a latency sensitivity metric;

generating from the network traffic data a bandwidth sensitivity metric; and

comparing the latency sensitivity metric and the bandwidth sensitivity metric to determine the computer network'"'"'s performance during operation of the software application.