System and method for simulating traffic loads in packetized communication networks

• US 7,885,803 B2
• Filed: 05/28/2003
• Issued: 02/08/2011
• Est. Priority Date: 05/28/2003
• Status: Expired due to Fees

• Current Assignee: Alcatel-Lucent USA, Inc. (Nokia Corporation)
• Original Assignee: Alcatel-Lucent USA, Inc. (Nokia Corporation)
• Inventors: Nuzman, Carl J., Weiss, Alan A., Saniee, Iraj
• Primary Examiner(s): Shah; Kamini S
• Assistant Examiner(s): Lo; Suzanne
• Application Number:
  US10/446,370
• Publication Number:
  US 20040240387A1
• Time in Patent Office: 2,813 Days
• Field of Search: 709/203, 709/232, 709/235, 370/230, 370/252, 370/412, 370/428, 703/21
• US Class Current: 703/21
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 41/24 using dedicated network management hardware

Abstract

A system for, and method of, simulating traffic loads in a packetized communication network. In one embodiment, the system includes: (1) an initial arrival generator that generates an initial arrival for a simulated user session at an independent time from a previous user session arrival and (2) a subsequent arrival generator, associated with the initial arrival generator, that generates a number of connections to be associated with the simulated user session as an independent variable and, if the number of connections is greater than one, generates a mean connection interarrival time for the simulated user session and a plurality of timings of subsequent arrivals.

Claims

1. A computer-readable medium having stored thereon a plurality of instructions including instructions that when executed by a processor cause said processor to implement a system for dynamically simulating transmission control protocol (TCP) traffic loads of types that may include, but are not limited to, Hypertext Transfer Protocol (HTTP) traffic in a wide-area network in user-defined epochs having a duration t which is less than a duration for a user session, said system comprising:

an initial arrival generator that generates an initial arrival for a simulated user session during an initial epoch at an independent time from a previous user session arrival in said wide-area network; and

a subsequent arrival generator that dynamically provides a prediction of a sequence of connection arrivals for subsequent epochs, said subsequent arrival generator associated with said initial arrival generator which generates a number of connections to be associated with said simulated user session during said initial epoch as an independent variable and, if said number of connections is greater than one, generates a mean connection interarrival time for said simulated user session during said initial epoch and a plurality of timings of subsequent connection arrivals.

2. The computer-readable medium as recited in claim 1 wherein said time is an exponentially distributed time.

3. The computer-readable medium as recited in claim 1 wherein said subsequent arrival generator generates said number of connections as an independent random biPareto variable.

4. The computer-readable medium as recited in claim 1 wherein said plurality of timings numbers one less than said number of connections.

5. The computer-readable medium as recited in claim 1 wherein said plurality of timings are independent Weibull random variables.

6. The computer-readable medium as recited in claim 1 wherein a number of said user-defined epochs is not determined prior to said user session.

7. The computer-readable medium as recited in claim 1 wherein at least one system parameter in one of said user-defined epochs differs from at least one system parameters in an other of said user-defined epochs.

8. A computer-implemented method of dynamically simulating transmission control protocol (TCP) traffic loads of types that may include, but are not limited to, Hypertext Transfer Protocol traffic in a wide-area network in user defined epochs having a duration t which is less than a duration for a user session, said method comprising:

generating an initial arrival for a simulated user session during an initial epoch at an independent time from a previous user session arrival in said wide-area network;

generating a number of connections to be associated with said simulated user session during said initial epoch as an independent variable; and

providing a dynamic prediction of a sequence of connection arrivals for subsequent epochs by generating a mean connection interarrival time for said simulated user session during said initial epoch and a plurality of timings of subsequent connection arrivals if said number of connections is greater than one.

9. The method as recited in claim 8 wherein said time is an exponentially distributed time.

10. The method as recited in claim 8 wherein said generating said number of connections comprises generating said number of connections as an independent random biPareto variable.

11. The method as recited in claim 8 wherein said plurality of timings numbers one less than said number of connections.

12. The method as recited in claim 8 wherein said plurality of timings are independent Weibull random variables.

13. The method as recited in claim 8 wherein said plurality of timings are scaled to conform to a period of time.

14. The method as recited in claim 8 further comprising generating subsequent arrivals based on a sorted list of said plurality of timings.

15. A computer-readable medium having stored thereon a plurality of instructions including instructions that when executed by a processor cause said processor to implement a system for dynamically simulating transmission control protocol (TCP) traffic loads of types that may include, but are not limited to, Hypertext Transfer Protocol (HTTP) traffic in a wide-area network during user defined epochs having a duration t which is less than a duration for a user session, said system comprising:

an initial arrival generator that generates initial arrivals for a simulated user session during an initial epoch at an independent, exponentially distributed time from previous user session arrivals wherein a rate of said user session arrivals corresponds to a wide-area network; and

a subsequent arrival generator that dynamically provides a sequence of connection arrivals for subsequent epochs, said subsequent arrival generator associated with said initial arrival generator which generates a number of connections to be associated with said simulated user sessions during said initial epoch as an independent variable and, if said number of connections is greater than one, generates a mean connection interarrival time for said simulated user sessions during said initial epoch and a plurality of timings of subsequent connection arrivals, said subsequent arrival generator generating subsequent arrivals based on a sorted list of said plurality of timings.

16. The computer-readable medium as recited in claim 15 wherein said subsequent arrival generator generates said number of connections as an independent random biPareto variable.

17. The computer-readable medium as recited in claim 15 wherein said plurality of timings numbers one less than said number of connections.

18. The computer-readable medium as recited in claim 15 wherein said plurality of timings of subsequent arrivals are generated based on a spatial aggregation of said previous user session arrivals.

19. The computer-readable medium as recited in claim 18 wherein said spatial aggregation is formed from a summation of multiple copies of an original trace of a user session whose rate of arrivals corresponds to a local area network, said copies generated by cutting said original trace at a random point and interchanging beginning and ending segments.

20. The computer-readable medium as recited in claim 15 wherein said system incorporates an algorithm that generates a point process of events for modeling Transmission Control Protocol (TCP) connections.