Method and apparatus for measuring the performance of a network based application program

• US 6,446,028 B1
• Filed: 11/25/1998
• Issued: 09/03/2002
• Est. Priority Date: 11/25/1998
• Status: Expired due to Term

• Current Assignee: Dynatrace LLC (Dynatrace, Inc.)
• Original Assignee: Keynote Systems (Dynatrace, Inc.)
• Inventors: Wang, George R.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): TSAI, CAROL S W
• Application Number:
  US09/199,972
• Time in Patent Office: 1,378 Days
• Field of Search: 702/122, 702/119, 702/176-178, 702/179, 702/182, 702/183, 702/186, 702/187, 702/188, 702/FOR.103, 702/FOR.104, 702/FOR.134, 702/FOR.135, 702/FOR.139, 702/FOR.170, 702/FOR.171, 709/223-226, 714/47, 714/51, 714/55, 370/229-237
• US Class Current: 702/186
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: H04L 41/508 based on type of value added network service under agreement
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0864 Round trip delays
• CPC Class Codes: H04L 43/106 using time related information in packets, e.g. by adding timestamps
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

A client-server software performance monitor system is disclosed. In the system of the present invention, a performance monitor machine is coupled to a computer network in close network proximity to one or more server systems that are to be monitored. The performance monitor machine monitors all network communication originating from or addressed to the server system to determine client-server transaction times. The performance monitor machine calculates a server processing time by subtracting the time when the server system receives the request packet from a client system from the time when the server system sends the first response packet back to the client system. The performance monitor machine also calculates an network transit time by subtracting said time when the server system receives the final acknowledgement packet from the time when said server system sent the first response packet to the client system. The performance monitor machine calculates an approximate total client observed response time by adding the server processing time and the approximate network transit time.

Claims

1. A method of monitoring the performance of a client-server application on a computer network, said method comprising:

coupling a performance monitor machine to said computer network in close network proximity to a first server system such that the performance monitor machine receives packets at approximately the same time that the first server system receives the packets;

sniffing from said performance monitor machine to determine a time when said first server system receives a request packet from a client system;

sniffing from said performance monitor machine to determine a time when said first server system sends a first response packet to said client system;

sniffing from said performance monitor machine to determine a time when said first server system receives a final acknowledgement packet from said client system; and

calculating client-server transaction times using said times of said sniffed packets.

2. The method as claimed in claim 1 wherein said step of calculating further comprises:

calculating a first server processing time by subtracting said time when said first server system receives said request packet from a client system from said time when said first server system sends said first response packet to said client system.

3. The method as claimed in claim 1 wherein said step of calculating further comprises:

calculating a network transit time by subtracting said time when said first server system receives said final acknowledgement packet from said client system from said time when said first server system sent said first response packet to said client system.

4. The method as claimed in claim 2 wherein said step of calculating further comprises:

calculating a network transit time by subtracting said time when said first server system receives said final acknowledgement packet from said client system away from said time when said first server system sent said first response packet to said client system.

5. The method as claimed in claim 4 wherein said step of calculating further comprises:

calculating a total round-trip response time by adding said network transit time to said first server processing time.

6. The method as claimed in claim 1 wherein said final acknowledgement packet comprises a TCP acknowledgement packet.

7. The method as claimed in claim 1 wherein said first server system is executing a SAP R/3 application.

8. The method as claimed in claim 1 further comprising:

sniffing from said performance monitor machine to determine a time when said first server system sends a request packet to a second server system;

sniffing from said performance monitor machine to determine a time when said first server system receives an acknowledgement packet from said second server system;

sniffing from said performance monitor machine to determine a time when said first server system receives a first response packet from said second server system;

sniffing from said performance monitor machine to determine a time when said first server system receives a final response packet from said second server system; and

calculating server-server transaction times using said times of said sniffed packets.

9. The method as claimed in claim 6 wherein calculating further comprises:

calculating a second server processing time by subtracting said time when said first server system receives said first response packet from said second server system from said time when said first server system receives said final response packet from said second server system.

10. The method as claimed in claim 8 wherein said step of calculating further comprises:

calculating a request round-trip transit time by subtracting said time when said first server system sends a request packet to a second server system from said time when said first server system receives an acknowledgement packet from said second server system;

calculating a total response receiving time by subtracting said time when said first server system receives a first response packet from said time when said first server system receives a final response packet from said second server system; and

calculating a second server transit time by adding said request round-trip transit time to said total response receiving time.

11. An arrangement for monitoring the performance of a client-server application, said apparatus comprising:

a computer network, said computer network for carrying packets between a plurality of client systems coupled to said computer network and at least one server system coupled to said computer network;

performance monitor machine, said performance monitor machine coupled to said computer network in close network proximity to a first server system such that the performance monitor machine receives packets at approximately the same time that the first server system receives the packets, said performance monitor machine sniffing packets originating from said first server system and received by said first server system.

12. The arrangement as claimed in claim 11 wherein said performance monitor machine calculates a first server processing time by subtracting a time when said first server system receives said request packet from a client system from a time when said first server system sends said first response packet to said client system.

13. The arrangement as claimed in claim 11 wherein said performance monitor machine calculates a network transit time by subtracting a time when said first server system receives a final acknowledgement packet from a client system from said time when said first server system sent said first response packet to said client system.

14. The arrangement as claimed in claim 12 wherein said performance monitor machine calculates a network transit time by subtracting said time when said first server system receives said final acknowledgement packet from said client system from said time when said first server system sent said first response packet to said client system.

15. The arrangement as claimed in claim 14 wherein said performance monitor machine calculates a total round-trip response time by adding said network transit time to said first server processing time.

16. The arrangement as claimed in claim 13 wherein said final acknowledgement packet comprises a TCP acknowledgement packet.

17. The arrangement as claimed in claim 11 wherein said first server system is executing a SAP R/3 application.

18. The arrangement as claimed in claim 11 wherein said performance monitor machine calculates a second server processing time by subtracting a time when said first server system receives a first response packet from said second server system in response to a request packet sent to said second server system from a time when said first server system receives said final response packet from said second server system.

19. The arrangement as claimed in claim 11 wherein said performance monitor machine calculates a request round-trip transit time by subtracting a time when said first server system sends a request packet to a second server system from said time when said first server system receives an acknowledgement packet from said second server system, a total response receiving time by subtracting a time when said first server system receives a first response packet from said second server system from a time when said first server system receives a final response packet from said second server system, and a second server transit time by adding said request round-trip transit time to said total response receiving time.

20. The arrangement as claimed in claim 18 wherein said performance monitor machine calculates a request round-trip transit time by subtracting a time when said first server system sends a request packet to a second server system from said time when said first server system receives an acknowledgement packet from said second server system, a total response receiving time by subtracting a time when said first server system receives a first response packet from said second server system from a time when said first server system receives a final response packet from said second server system, and a second server transit time by adding said request round-trip transit time to said total response receiving time.

22. The arrangement as claimed in claim 11 further comprising:

a database first server coupled to said computer network.

21. An arrangement for monitoring the performance of a client-server application, said apparatus comprising:

a computer network, said computer network for carrying packets between a plurality of client systems coupled to said computer network and at least one server system coupled to said computer network;

performance monitor machine, said performance monitor machine coupled said computer network in close network proximity to a first server system, said performance monitor machine sniffing packets originating from said first server system and received by said first server system, said performance monitor machine configured to calculate a request round-trip transit time by subtracting a time when said first server system sends a request packet to a second server system from said time when said first server system receives an acknowledgement packet from said second server system, a total response receiving time by subtracting a time when said first server system receives a first response packet from said second server system from a time when said first server system receives a final response packet from said second server system, a second server transit time by adding said request round-trip transit time to said total response receiving time, a second server processing time by subtracting a time when said first server system receives a first response packet from said second server system in response to a request packet sent to said second server system from a time when said first server system receives said final response packet from said second server system, and a total second server response time by adding said second server processing time to said second server transit time.