Expert system method for performing window protocol-based data flow analysis within a data communication network

• US 5,197,127 A
• Filed: 09/24/1990
• Issued: 03/23/1993
• Est. Priority Date: 09/24/1990
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Springsteen, Robert H., Waclawsky, John G., Daugherty, Raymond F.
• Primary Examiner(s): RICHARDSON, ROBERT
• Application Number:
  US07/586,828
• Time in Patent Office: 911 Days
• Field of Search: 364/DIG. 1, 364/DIG. 2, 395/200, 370/94.1, 370/60
• US Class Current: 709/224
• CPC Class Codes: G06F 13/00 Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units interface circuits for specific input/output devices G06F3/00 ; multiprogram control therefor G06F9/46; multiprocessor systems G06F15/16
• CPC Class Codes: Y10S 706/917 Communication

Abstract

An expert system method is disclosed for analyzing window protocol-based data flows in a data communications network. The method includes the use of a data flow efficiency state variable S which is a binary number having at least three bits which reflect the number of packets transmitted by a node, the queuing of packets within a node, and the congestion of the packets within the node or in the node connected to the node of interest. After having assigned a value to the state variable, a knowledge base is accessed containing network problem determination recommendations for optimizing data flow efficiency within the network.

Claims

1. An expert system method for analysing the window protocol-based data flow in a data communications network over which data packets are transmitted, said data packets including a data portion and a header portion, comprising:

setting a packet transmission window to have a maximum quantity of N packets which can be transmitted within an interval from a node in the network;

setting a queued packet threshold value to a quantity of C packets which may be held in a queue during an interval at the node;

defining a data flow efficiency state variable S as a binary number having at least three bits, with a first bit B1;

counting the number of packets transmitted from the node during a measurement interval and setting B1 if the number of packets transmitted is equal to N;

setting B2 if any packet is held in the queue during the measurement interval;

setting B3 if more than C packets are held in the queue during the measurement interval;

determining the value of said data flow efficiency state variable S from values of B1, B2 and B3 set by said counting and setting steps;

accessing a knowledge base containing network problem determination recommendations which are accessible with said value of said data flow efficiency state variable S;

outputting a problem determination recommendation for optimizing data flow efficiency in said network in response to accessing said knowledge base with said value of S.

2. The method of claim 1, wherein the step of determining the value of the data flow efficiency state variable S further comprises the steps of:

establishing an initial state assignment for a measurement interval; and

comparing the number of congested intervals to the total number of held intervals.

3. The method of claim 1, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the byte count during the measurement interval exceeds a predetermined value.

4. The method of claim 1, wherein the step of determining the value of the data flow efficiency stat variable S further comprises the step of:

determining whether the number of held intervals exceeds a predetermined percentage of total transmission intervals.

5. The method of claim 1, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the number of congested intervals exceeds a predetermined percentage of the total number of transmission intervals in the measurement interval.

6. The method of claim 1, wherein the step of determining the value of the data flow efficiency state variable S further comprises the steps of:

providing a second queue threshold value which is greater than said first threshold value; and

determining whether said second threshold value is exceeded for any transmission interval in said measurement interval.

7. The method of claim 1 wherein said step of accessing a knowledge base containing network problem determination recommendations further comprising the steps of:

evaluating the most significant value of the data flow efficiency state variable; and

combining said most significant state value with path configuration information for said node.

8. An expert system method for analyzing the window protocol-based data flow in a data communications network over which data packets are transmitted, said data packets including a data portion and a header portion, comprising:

setting a packet transmission window to have a maximum quantity of N packets which can be transmitted within an interval from a terminal in the network;

setting a queued packet threshold value to a quantity of C packets which may be held in a queue during an interval at the terminal;

defining a data flow efficiency state variable S as a binary number having at least three bits, with a first bit B1, a second bit B2 and a third bit B3;

receiving at the terminal during a first interval, a plurality of Ml packets which is less than N packets from a source input;

transmitting from the terminal during a second interval said M1 packets over said network to a destination;

receiving at the terminal during said second interval, a plurality of M2 packets which is more than N packets from said source input;

transmitting from said terminal during a third interval, N of said M1 packets over said network to said destination and queuing M2-N packets at said terminal;

marking during said third interval an N+1st packet of said queued packets as being held in said queue;

marking during said third interval said N+1st packet of the queued packets as being congested in said queue if there are more than C packets in said queue;

counting the number of packets transmitted from the terminal during a measurement period including a plurality of intervals and setting B1 if the number of packets transmitted during an interval is equal to N;

setting B2 if any packet is held in the queue during the measurement period;

setting B3 if more than C packets are held in the queue during any interval in the measurement period;

determining the value of said data flow efficiency state variable S from values of B1, B2 and B3 set by said counting and setting steps;

accessing a knowledge base containing network problem determination recommendations which are accessible with said value of said data flow efficiency state variable S;

outputting a problem determination recommendation for optimizing data flow efficiency in said network in response to accessing said knowledge base with said value of S.

9. The method of claim 8, wherein the step of determining the value of the data flow efficiency state variable S further comprises the steps of:

establishing an initial state assignment for a measurement period; and

comparing the number of congested intervals to the total number of held intervals.

10. The method of claim 8, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the byte count during the measurement period exceeds a predetermined value.

11. The method of claim 8, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the number of held intervals exceeds a predetermined percentage of total intervals.

12. The method of claim 8, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the number of congested intervals exceeds a predetermined percentage of the total number of intervals in the measurement interval.

13. The method of claim 8, wherein the step of determining the value of the data flow efficiency state variable S further comprises the steps of:

providing a second queue threshold value which is greater than said first said threshold value; and

determining whether said second threshold value is exceeded for any transmission interval in said measurement interval.

14. The method of claim 8, wherein said step of accessing a knowledge base containing network problem determination recommendations further comprising the steps of:

evaluating the most significant value of the data flow efficiency state variable; and

combining said most significant state value with configuration information for said terminal.

15. An expert system method for analysing the window protocol-based data flows in a data communications network over which plural data packets are transmitted during a transmission, said analyzing being performed over a measurement interval including a plurality of said transmission intervals, said data packets including a data portion and a header portion, comprising:

setting a packet transmission window to have a maximum quantity of N packets which can be transmitted within a transmission interval from a node in the network;

setting a queued packet threshold value to a quantity of C packets which may be held in a queue during a transmission interval at the node;

defining a data flow efficiency state variable S as a binary number having at least three bits, with a first bit B1, a second bit B2 and a third bit B3;

counting the number of packets transmitted from the node during a measurement interval as a packet count and setting B1 if the number of packets transmitted in any transmission interval is equal to N;

setting B2 if any packet is held in the queue during the measurement interval;

setting B3 if more than C packets are held in the queue during any transmission interval in the measurement interval;

determining an initial value of said data flow efficiency state variable S from values of B1, B2 and B3 set by said counting and setting steps;

selectively modifying said value of said data flow efficiency state variable S to a modified value based on said packet count;

accessing a knowledge base containing network problem determination recommendations which are accessible with said modified value of said data flow efficiency state variable S;

outputting a problem determination recommendation for optimizing data flow efficiency in said network in response to accessing said knowledge base with said modified value of S.

16. The method of claim 15, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

comparing the number of congested intervals to the total number of held intervals.

17. The method of claim 15, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the byte count during the measurement interval exceeds a predetermined value.

18. The method of claim 15, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the number of held intervals exceeds a predetermined percentage of total transmission intervals.

19. The method of claim 15, wherein the step of determining the value of the data flow efficiency state variable S further comprises the step of:

determining whether the number of congested intervals exceeds a predetermined percentage of the total number of transmission intervals in the measurement interval.

20. The method of claim 15, wherein the step of determining the value of the data flow efficiency state variable S further comprises the steps of:

providing a second queue threshold value which is greater than said first said threshold value; and

determining whether said second threshold value is exceeded for any transmission interval in said measurement interval.

21. The method of claim 15, wherein said step of accessing a knowledge base containing network problem determination recommendations further comprising the steps of:

evaluating the most significant value of the data flow efficiency state variable; and

combining said most significant state value with configuration information for said node.