Quality of service in IT infrastructures

• US 7,924,732 B2
• Filed: 04/19/2005
• Issued: 04/12/2011
• Est. Priority Date: 04/19/2005
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Tischhaeuser, Martin, Schimmelpfeng, Joern, Nitsch, Frank
• Primary Examiner(s): Ngo; Ricky
• Assistant Examiner(s): Liu; Ben H
• Application Number:
  US11/109,285
• Publication Number:
  US 20060245369A1
• Time in Patent Office: 2,184 Days
• Field of Search: 370/230, 370/231, 370232-235, 370252-253, 370/229
• US Class Current: 370/252
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 43/0817 by checking functioning

Abstract

A method of determining quality of service (QoS) in an IT infrastructure on the basis of QoS-related data received from data sources for these QoS-related data, wherein the data from the data sources regularly arrive in real-time, but data from one or more of the data sources may occasionally be delayed. On the basis of the regular data, a real-time indication of the quality of service is calculated in a first process. This indication is possibly inaccurate since delayed data is not included. After the receipt of delayed data, a delayed, but more accurate indication of the quality of service is calculated in a second concurrent process on the basis of regular information, which comprises at least one of regular data and information derived from it, and the delayed data.

Claims

1. A method of determining quality of service (QoS) in an IT infrastructure on the basis of QoS-related data received from data sources for these QoS-related data, wherein the data from the data sources regularly arrive in real-time, but data from one or more of the data sources may occasionally be delayed, the method comprising:

a QoS management computer system calculating, on the basis of the regular data, a real-time indication of the quality of service in a first process for a calculation cycle, said indication not being based on delayed data;

the QoS management computer system calculating, after the receipt of delayed data, on the basis of regular information, which comprises at least one of regular data and information derived from it, and the delayed data, a delayed, but more accurate indication of the quality of service in a second concurrent process for the same calculation cycle.

2. The method of claim 1, wherein the activities of calculating comprise calculating QoS parameter values for short time-intervals, thus forming calculation cycles, and combining these QoS parameter values over time.

3. The method of claim 2, wherein the data sources provide their QoS-related data with a timestamp indicating the point of time when the data was generated, a data and its timestamp being called “

datapoint”

, and wherein, at least in the second process, the datapoints are put in correct order and are combined based on their timestamps.

4. The method of claim 3, wherein, at least in the second process, only datapoints are combined in the calculation cycle which, due to their timestamp, belong to said calculation cycle.

5. The method of claim 3, wherein the timestamp of a delayed datapoint is incremented when the datapoint has been generated between the occurrence of the delay and its detection, so that the datapoint is used in the second process as if it had been generated after the detection of the delay.

6. The method of claim 2, wherein the calculation of QoS parameter values in the calculation cycles is based on a calculation graph indicating how the data from the data sources are to be combined.

7. The method of claim 6, wherein the second process is based on the complete calculation graph, or a part of the calculation graph, used in the first process.

8. The method of claim 6, wherein the first process reveals intermediate results, which are represented by nodes of the calculation graph, wherein some of the intermediate results are affected by a data source'"'"'s non-availability, while other intermediate results are not affected by the data source'"'"'s non-availability and are therefore correctly calculated in the first process,

wherein non-affected intermediate results which influence other nodes in the calculation graph are buffered and are re-used in the second process, which is therefore based on a different calculation graph.

9. The method of claim 6, wherein the processes are arranged to handle schedule and configuration events which trigger the use of modified calculation graphs, wherein the modified calculation graphs are used in both the first and second process in a time-consistent manner.

10. The method of claim 1, wherein after the second process has caught up with the first process, the first process is synchronized with the second process, and the second process is suspended.

11. The method of claim 1, wherein after the second process has caught up with the first process, the second process replaces the first process.

12. The method of claim 1, wherein the data received from a data source are data the receipt of which is expected, since they are data periodically sent by the data source, or upon request from the QoS management system.

13. The method of claim 12, wherein the availability of a data source is ascertained by an inference mechanism and wherein, in response to a detected non-availability, the second process is triggered.

14. The method of claim 1, wherein the data received from a data source are event data asynchronously transmitted.

15. The method of claim 14, wherein a data sources is polled from time to time and thereby the availability of the data source is ascertained, and wherein, in response to a detected non-availability, the second process is triggered.

16. The method of claim 14, wherein, when data from a data source had been delayed, but could finally be transmitted, the data source sends an end-of-transmission indication that all delayed data have been transmitted.

17. A computer system for determining quality of service (QoS) in an IT infrastructure on the basis of QoS related data received from data sources for these QoS related data, wherein the data from the data sources regularly arrive in real-time, but data from one or more of the data sources may occasionally be delayed, the computer system being programmed to:

calculate, on the basis of the regular data, a real-time indication of the quality of service in a first process for a calculation cycle, said indication not being based on delayed data;

calculate, after the receipt of delayed data, on the basis of regular information, which comprises at least one of regular data and information derived from it, and the delayed data, a delayed, but more accurate indication of the quality of service in a second concurrent process for the same calculation process.

18. A computer program product which is in the form of a machine-readable medium with program code stored on it,

wherein the program code is arranged to carry out a method, when executed on a computer system, of determining quality of service (QoS) in an IT infrastructure on the basis of QoS related data received from data sources for these QoS related data, wherein the data from the data sources regularly arrive in real-time, but data from one or more of the data sources may occasionally be delayed, said method comprising:

calculating, on the basis of the regular data, a real-time indication of the quality of service in a first process for a calculation cycle, said indication not being based on delayed data;

calculating, after the receipt of delayed data, on the basis of regular information, which comprises at least one of regular data and information derived from it, and the delayed data, a delayed, but more accurate indication of the quality of service in a second concurrent process for the same calculation cycle.