Method and system for real-time burstiness analysis of network traffic

• US 5,343,465 A
• Filed: 06/11/1993
• Issued: 08/30/1994
• Est. Priority Date: 06/11/1993
• Status: Expired due to Term

• Current Assignee: TTI Inventions A LLC (Intellectual Ventures LLC)
• Original Assignee: Bell Communications Research, Inc. (Telefonaktiebolaget LM Ericsson)
• Inventors: Khalil, Khalid M.
• Primary Examiner(s): Chin, Wellington
• Assistant Examiner(s): MARCELO, MELVIN C
• Application Number:
  US08/075,252
• Time in Patent Office: 445 Days
• Field of Search: 370/17, 370/60, 370/94.1, 370/13, 370/79, 371/20.1
• US Class Current: 370/232
• CPC Class Codes: H04L 12/2854 Wide area networks, e.g. public data networks
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks

Abstract

Method, for use with a communications network, for measuring and analyzing the burstiness of network traffic based on the ratio of packet interarrival times. The network traffic includes at least one data burst, and each data burst includes a stream of data packets for travel over the network. The method includes (a) receiving a first data packet of the at least one data burst for travel over the network, the first data packet having an associated first packet interarrival time, and (b) consecutively receiving a second data packet for travel over the network, the second data packet having an associated second packet interarrival time. The method also includes (c) determining the ratio of the second packet interarrival time to the first packet interarrival time and (d) comparing the ratio of the second packet interarrival time and first packet interarrival time to at least one predetermined constant. Lastly, the method includes (e) determining whether the second data packet belongs to the at least one data burst based on the comparison, and (f) accumulating the second data packet with a data burst based on the comparison. A system for carrying out the method is also provided.

Claims

1. A method, for use with a communications network, for measuring and analyzing the burstiness of network traffic which includes at least one data burst, each data burst including a plurality of data packets travelling over the network, each data packet having an associated interarrival time upon arrival at the network, the method comprising:

determining a ratio of a first packet interarrival time and a second packet interarrival time, the first packet interarrival time being associated with a first data packet belonging to the at least one data burst, the second packet interarrival time being associated with a second data packet, the data packets consecutively arriving at the network;

comparing the ratio of the interarrival times to at least one predetermined constant; and

determining whether the second data packet belongs to the at least one data burst based on the comparison.

2. The method of claim 1 further comprising accumulating the second data packet with a data burst based on the comparison.

3. The method of claim 2 wherein the ratio of the interarrival times is represented by R and the at least one predetermined constant is represented by C, the method further comprising accumulating the second data packet with the at least one data burst if 1/C≦

R ≦

C.

4. A method, for use with a communications network, for measuring and analyzing the burstiness of network traffic which includes at least one data burst, each data burst including a stream of data packets travelling over the network, the method comprising:

(a) receiving a first data packet of the at least one data burst for travel over the network, the first data packet having an associated first packet interarrival time;

(b) consecutively receiving a second data packet for travel over the network, the second data packet having an associated second packet interarrival time;

(c) determining the ratio of the second packet interarrival time to the first packet interarrival time;

(d) comparing the ratio of the second packet interarrival time and first packet interarrival time to at least one predetermined constant;

(e) determining whether the second data packet belongs to the at least one data burst based on the comparison; and

(f) accumulating the second data packet with a data burst based on the comparison.

5. The method of claim 4 further comprising:

(g) repeating steps (a)-(f) at the arrival of data packets so as to group the stream of data packets into data bursts; and

(h) constructing a burst profile of the network traffic as a series of data bursts rather than a series of packets.

6. The method of claim 4 wherein the ratio of the second packet interarrival time and the first packet interarrival time is represented by R and the at least one predetermined constant is represented by C, the method further comprising accumulating the second data-packet with the at least one data burst if 1/C≦

R ≦

C.

7. The method of claim 4 wherein the step of determining the ratio comprises:

determining the ratio of the second packet interarrival time to the average of n previous packet interarrival times of the at least one data burst so as to reduce the probability of erroneously grouping the second data packet with the at least one data burst.

8. The method of claim 7 further comprising:

(g) repeating steps (a)-(f) at the arrival of data packets so as to group the stream of data packets into data bursts; and

(h) constructing a burst profile of the network traffic as a series of data bursts rather than a series of packets.

9. The method of claim 8 wherein the ratio of the second packet interarrival time and the first packet interarrival time is represented by R and the at least one predetermined constant is represented by C, the method further comprising accumulating the second data packet with the at least one data burst if 1/C≦

S R ≦

C.

10. The method of claim 4 wherein the communications network is a frame relay network.