Metric transport and database load

• US 7,617,313 B1
• Filed: 12/27/2004
• Issued: 11/10/2009
• Est. Priority Date: 12/27/2004
• Status: Active Grant

• Current Assignee: T-Mobile Innovations LLC (Deutsche Telekom AG)
• Original Assignee: Sprint Communications Company LP (Deutsche Telekom AG)
• Inventors: Farrell, Brian Paul, Washburn, Brian John
• Primary Examiner(s): Vaughn, Jr.; William C
• Assistant Examiner(s): Christensen; Scott
• Application Number:
  US11/022,955
• Time in Patent Office: 1,779 Days
• Field of Search: 709/224
• US Class Current: 709/224
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/87 Monitoring of transactions

Abstract

Systems and methods for collecting metrics from monitored systems are provided. Collected metrics may be further used for modeling system performance. Collected metrics may be analyzed to identify correlations between metrics, which may then be a basis for system modeling. One or more alarm thresholds may be set based upon system models. If the monitoring of a system indicates that an alarm threshold has been passed, the appropriate alarm may be issued. Alarm levels may vary for different alarm thresholds.

Claims

1. A method for monitoring system metrics in a production environment, the method comprising:

measuring system metrics comprising at least a system throughput and a system load from a plurality of systems in the production environment;

reporting at least the system throughput and the system load to a monitoring computer;

storing at least the system throughput and the system load by the monitoring computer;

exporting at least the system throughput and the system load from the monitoring computer to a database;

analyzing at least the system throughput and the system load in the database to identify correlations between the system metrics;

modeling the system throughput as a function of system load by fitting a non-linear curve to the system throughput and system load to obtain a non-linear model of at least the system throughput as a function of the system load;

forecasting system performance in a production environment using the modeling of at least the system throughput as a function of the system load;

identifying a threshold for the system throughput at which system performance becomes unacceptable;

measuring the system throughput against the identified threshold; and

triggering an alarm to a system administrator when the system throughput exceeds the identified threshold.

2. The method for collecting system metrics of claim 1, wherein reporting at least the system throughput and the system load to a monitoring computer further comprises reporting at least the system throughput and the system load in a first format.

3. The method for collecting system metrics of claim 2, wherein exporting at least the system throughput and the system load from the monitoring computer to database further comprises exporting at least the system throughput and the system load in a second format.

4. The method for collecting system metrics of claim 3, further comprising:

converting by the monitoring computer, at least the system throughput and the system load reported to the monitoring computer in the first format to the second format.

5. The method for collecting system metrics of claim 4, wherein the first format comprises a propriety format.

6. The method for collecting system metrics of claim 5, wherein the second format comprises a non-proprietary format.

7. A system for monitoring system metrics in a production environment, the system comprising:

system monitoring software operating on a system in a production environment to be monitored, the system monitoring software measuring system metrics comprising at least a system throughput and a system load;

a monitoring computer operably connected to the system monitoring software over a network, the monitoring computer collecting at least the system throughput and the system load from the system monitoring software over the network operably connecting the system monitoring software to the monitoring computer;

a database communicatively linked to the monitoring computer, the database receiving at least the system throughput and the system load from the monitoring computer and storing at least the system throughput and the system load;

an analysis system that identifies correlations between at least the system throughput and the system load stored in the database and models the system throughput as a function of system load by fitting a non-linear curve to the system throughput and system load to obtain a non-linear model of at least the system throughput as a function of the system load,

wherein the modeling of the system metrics forecasts system performance in a production environment using the modeling of at least the system throughput as a function of the system load identifies a threshold at which system performance becomes unacceptable, and

wherein the analysis system measures the system throughput against the identified threshold and triggers an alarm to a system administrator when the system throughput exceeds the identified threshold.

8. The system for collecting system metrics of claim 7, wherein the system monitoring software measures at least the system throughput and the system load and places them in a first file format.

9. The system for collecting system metrics of claim 8, wherein the database stores at least the system throughput and the system load in a second file format.

10. The system for collecting system metrics of claim 9, wherein the monitoring computer converts at least the system throughput and the system load in a first file format to a second file format.

11. The system for collecting system metrics of claim 10, wherein the system monitoring software comprises proprietary software.

12. A computer readable storage media having stored thereon computer readable code that, when executed by a processor, causes a computer to perform a method for monitoring system metrics in a production environment, the method comprising:

measuring system metrics comprising at least a system throughput and a system load from a plurality of systems in the production environment;

reporting at least the system throughput and the system load to a monitoring computer;

storing at least the system throughput and the system load by the monitoring computer;

exporting at least the system throughput and the system load from the monitoring computer to a database;

analyzing at least the system throughput and the system load in the database to identify correlations between the system metrics;

modeling the system throughput as a function of system load by fitting a non-linear curve to the system throughput and system load to obtain a non-linear model of at least the system throughput as a function of the system load;

forecasting system performance in a production environment using the modeling of at least the system throughput as a function of the system load;

identifying a threshold for the system throughput at which system performance becomes unacceptable;

measuring the system throughput against the identified threshold; and

triggering an alarm to a system administrator when the system throughput exceeds the identified threshold.

13. The computer readable storage media of claim 12, wherein reporting at least the system throughput and the system load to a monitoring computer further comprises reporting at least the system throughput and the system load in a first format.

14. The computer readable storage media of claim 13, wherein exporting at least the system throughput and the system load from the monitoring computer to database further comprises exporting at least the system throughput and the system load in a second format.

15. The computer readable storage media of claim 14, wherein the method further comprises:

converting by the monitoring computer, at least the system throughput and the system load reported to the monitoring computer in the first format to the second format.

16. The computer readable storage media of claim 15, wherein the first format comprises a propriety format.

17. The computer readable storage media of claim 15, wherein the second format comprises a non-proprietary format.