System and method for analysis of communications networks

• US 7,769,850 B2
• Filed: 12/23/2004
• Issued: 08/03/2010
• Est. Priority Date: 12/23/2004
• Status: Expired due to Fees

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Silverman, Robert M.
• Primary Examiner(s): Vaughn, Jr.; William C
• Assistant Examiner(s): Gupta; Muktesh G
• Application Number:
  US11/019,453
• Publication Number:
  US 20060153089A1
• Time in Patent Office: 2,049 Days
• Field of Search: 709/223, 709/224, 709/242, 370/252, 370/254, 707/3
• US Class Current: 709/224
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route

Abstract

A system and method is provided for using network asset management data including attributes such as device type, location, and link speed for modeling network performance and identifying performance issues. The invention provides measured and observed performance data for one or more segments of a network (e.g., one or more hops or end to end) so that identification of diminished performance in the network might be accomplished. Additionally, the invention may provide projections of performance impacts based upon proposed changes to one or more components in the network. Performance metrics, ping times, MIB data, link speeds, propagation delays, device latency, serialization rates, conversational speeds, conversational utilization and other data are used to model a network for determining or isolating problematic issues and to forecast performance impacts of potential changes to the network.

Claims

1. A method of modeling a network, comprising the steps of:

a server which acquires one or more metrics associated with one or more network hops, wherein the one or more metrics are derived from management information base (MIB) data, which reflects a network state device-by-device and port-by-port across the network over user-specified time intervals, the MIB data is configured to be merged with ping data and analyzed for inaccuracies, the ping data is configured to be used to overcome the inaccuracies in the MIB data;

the server which calculates one or more performance ratings from the one or more metrics for the one or more network hops;

the server which computes an overall baseline capability of the one or more network hops based on the calculating step for use in at least any one of characterizing impacts of one or more applications, identifying diminished performance by a network component and projecting an effect caused by altering the network component; and

utilizing the server to create a database model of the network'"'"'s connectivity and to determine baseline and current performance level of the network,

wherein the ping data comprises longer and shorter ping packets transmitted in isolation from each other and to provide data for analysis of the baseline performance level of the network.

2. The method of claim 1, wherein the acquiring step comprises at least any one of the following steps:

determining a link speed associated with the one or more hops;

determining a propagation delay associated with the one or more hops;

determining serialization rate; and

determining a device latency associated with the one or more hops.

3. The method of claim 1, wherein the acquiring step comprises at least any one of the following steps:

determining a utilization ratio associated with the one or more hops;

determining a byte rate in associated with the one or more hops;

determining a byte rate out associated with the one or more hops;

determining a packet rate in associated with the one or more hops;

acquiring the MIB data;

acquiring trace information; and

determining a ping time associated with the one or more hops.

4. The method of claim 1, wherein the calculating step includes performing a single hop analysis based on link speed, propagation delay, device latency and utilization.

5. The method of claim 4, further comprising performing a multi-hop analysis in a multi-hop network based on more than one single hop analysis.

6. The method of claim 5, further comprising calculating a packet handling over distance (PHOD) for the multi-hop network.

7. The method of claim 1, wherein computing the overall baseline includes computing a conversational utilization of the one or more network hops based at least on conversational speed and serialization time.

8. The method of claim 1, wherein computing the overall baseline includes computing a streaming utilization based at least on a streaming speed and a currently available network speed.

9. The method of claim 1, wherein the acquiring step acquires the one or more metrics for one or more network devices associated with the one or more network hops.

10. The method of claim 1, further comprising providing a projected performance model based on changing a parameter associated with the one or more metrics for identifying potential impacts of the changed parameter.

27. The method of claim 1, further comprising triggering a call to a provider of edge network services when a network component response time is sub-par for an industry.

28. The method of claim 1, further comprising using the one or more metrics to determine where and when to invest in infrastructure.

29. The method of claim 1, further comprising employing application profiling to determine traffic characteristics presented to the network by the one or more applications.

30. The method of claim 29, wherein the application profiling is derived from traces, application traffic flow analyzing software, application documentation, and application designers.

31. The method of claim 1, further comprising what-if modeling of potential changes to the network, including effects on performance of additional traffic, link speed increases, and changes in distance between clients and servers.

32. The method of claim 31, wherein the one or more metrics include speed associated with a hop which includes theoretical and parametrically controlled speeds, propagation delays, device latency, utilization, packet loss rate, byte rate in, byte rate out, packet rate in, packet rate out, and ping time across hop.

33. The method of claim 32, further comprising application profiling employed to determine characteristics of traffic presented to the network by applications, wherein the profiling includes profiling applications by sets of units of work, each of which requires a specific number of bytes to be transmitted and received in a specific number of trips across the network for completion.

34. The method of claim 33, wherein the units of work include conversational units of work comprising a single transmission of “

m”

bytes and a single receipt of “

n”

bytes;

multi-turnaround conversational units of work comprising “

p”

turnarounds with a total of “

m”

bytes transmitted and “

n”

bytes received (e.g., a SQL-based retrieval of data base rows using acknowledge and “

get-next”

function) or streaming units of work consisting of “

n”

megabytes of data to be transmitted.

35. The method of claim 1, wherein the method identifies for modeling one or more network paths and further comprises:

computing latency of the network along the one or more network paths;

computing a streaming utilization along the one or more network paths;

utilizing what-if modeling to ascertain potential impacts on performance along the one or more network paths; and

comparing the performance with the baseline performance level of the network.

11. A method for managing a network, comprising the steps of:

a server which identifies one or more network paths to be analyzed;

collecting management information base (MIB) data and ping data from one or more network devices along the one or more network paths, the MIB data configured to reflect a network state device-by-device and port-by-port across the network over user-specified time intervals, the MIB data also configured to be merged with the ping data and analyzed for inaccuracies, which are overcome by the ping data;

the server which computes one or more of the following outputs from the MIB and the ping data;

(i) a baseline and currently available streaming speed for the one or more network paths;

(ii) a currently available conversational speed of the one or more network paths;

(iii) a latency of the one or more network paths;

(iv) a current streaming utilization of the one or more network paths;

(v) a current queue time along the one or more network paths;

the server which generates from the one or more outputs of the computing step for use in identifying diminished performance along the one or more network paths, one or more of the following;

(i) a baseline and current performance of streaming applications, and

(ii) a baseline and current performance of conversational applications; and

utilizing the server to create a database model of the network'"'"'s connectivity and to determine baseline and current performance level of the network,

wherein the ping data comprises longer and shorter ping packets transmitted in isolation from each other and to provide data for analysis of the baseline performance level of the network.

12. The method of claim 11, further comprising tuning streaming applications under baseline and current conditions.

13. The method of claim 11, further comprising:

changing one or more parameters associated with at least any one of the MIB data, the ping data and a user specified parameter; and

projecting new one or more outputs by repeating the computing and generating steps.

14. A system in combination with hardware for modeling a network, comprising:

means for acquiring metrics associated with a portion of a network, wherein the metrics are derived from management information base (MIB) data, which reflects a network state device-by-device and port-by-port across the network over user-specified time intervals, the MIB data is configured to be merged with ping data and analyzed for inaccuracies, the ping data is configured to be used to overcome the inaccuracies in the MIB data;

means for calculating one or more performance ratings from the one or more metrics;

means for computing an outputting an overall baseline capability of the portion of the network based on the performance ratings for use in at least any one of characterizing impacts of one or more applications, identifying diminished performance by a network component and projecting an effect caused by altering a network component;

a database model modeling the network'"'"'s connectivity; and

a system for determining baseline and current performance level of the network,

wherein the ping data comprises longer and shorter ping packets transmitted in isolation from each other and to provide data for analysis of the baseline performance level of the network.

15. The system of claim 14, further comprising at least any one of the following:

means for determining a link speed associated with the portion of the network;

means for determining a propagation delay associated with the portion of the network;

means for determining a serialization rate for the portion of the network; and

means for determining latency associated with the portion of the network.

16. The system of claim 14, further comprising at least any one of the following:

means for determining a utilization ratio associated with the portion of the network;

means for determining a byte rate in associated with the portion of the network;

means for determining a byte rate out associated with the portion of the network;

means for determining a packet rate in associated with the portion of the network;

means for acquiring the MIB data;

means for acquiring trace information; and

means for determining a ping time associated with the portion of the network.

17. The system of claim 14, further comprising means for performing a single hop analysis based on link speed, propagation delay, device latency and utilization.

18. The system of claim 14, further comprising a means for performing a multi-hop analysis in a multi-hop network based on more than one single hop analysis.

19. The system of claim 14, further comprising a means for calculating a packet handling over distance (PHOD) for the multi-hop network.

20. The system of claim 14, further comprising means for computing a conversational utilization of the portion of the network based at least on conversational speed and serialization time.

21. The system of claim 14, further including a means for computing a streaming utilization based at least on a streaming speed and a currently available network speed.

22. The system of claim 14, further comprising a means for acquiring the one or more metrics for one or more network devices associated with the portion of the network.

23. The system of claim 14, further comprising a means for providing a projected performance model based on changing a parameter associated with the one or more metrics for identifying potential impacts of the changed parameter.

24. The system of claim 14, wherein the portion of the network is one of a single hop, a multi-hop and an end to end path.

25. The system of claim 14, further comprising a means to display the overall baseline capability.

26. A computer program product comprising a computer usable storage medium having readable program code embodied in the storage medium, the computer program product being configured to model a network and including at least one component to:

identify one or more network paths to be analyzed;

collect management information base (MIB) data and ping data from one or more network devices along the one or more network paths, the MIB data configured to reflect a network state device-by-device and port-by-port across the network over user-specified time intervals, the MIB data also configured to be merged with the ping data and analyzed for inaccuracies, which are overcome by the ping data;

compute one or more of the following outputs from the MIB and the ping data for tuning or identifying problems in the one or more network paths

(i) a baseline and currently available streaming speed for the one or more network paths;

(ii) a currently available conversational speed of the one or more network paths;

(iii) a latency of the one or more network paths;

(iv) a current streaming utilization of the one or more network paths;

(v) a current queue time along the one or more network paths; and

generate from the one or more outputs of the computing step for use in identifying diminished performance along the one or more network paths, one or more of the following;

(i) a baseline and current performance of streaming applications, and

(ii) a baseline and current performance of conversational applications;

create a database model of the network'"'"'s connectivity; and

determine baseline and current performance levels of the network,

wherein the ping data comprises longer and shorter ping packets transmitted in isolation from each other and to provide data for analysis of the baseline performance level of the network.