Performance monitoring

• US 9,602,340 B2
• Filed: 04/20/2007
• Issued: 03/21/2017
• Est. Priority Date: 04/20/2007
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Kroll, Thomas, Scherf, Ina
• Primary Examiner(s): Wang, Liangche A
• Assistant Examiner(s): LIN, SHERMAN L
• Application Number:
  US11/737,919
• Publication Number:
  US 20080263195A1
• Time in Patent Office: 3,623 Days
• Field of Search: 709/224, 709/223
• US Class Current: 1/1
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: H04L 41/064 involving time analysis
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks

Abstract

Real-time data is extracted from a log file for at least one monitored process on each of a plurality of servers. For each monitored process, extracted real-time data is inserted into object variables of an object. The object variable data is processed to generate performance statistics for the monitored processes and to determine whether to trigger an alarm. A database is updated with object variable data and performance statistics when an event associated with a monitored process takes place. Historical performance statistics from the database are compared with current performance statistics to determine a performance trend.

Claims

1. A method, comprising

extracting real-time data from a log file for at least one monitored process on each of the plurality of child servers with a corresponding log file reader at a parent server;

iteratively reading command process information from the log files with the corresponding to file readers, each command process information contained in a single line, and determining how to send the single line to the parent server until the single line indicates that a corresponding child server has shut down or until the single line cannot be read;

for each monitored process, inserting extracted real-time data into object variables of an object;

processing object variable data to generate performance statistics for the monitored processes and to determine whether to trigger an alarm, the processing including identifying how many monitored processes are running in parallel and which of the running monitored processes is the oldest to identify one or more monitored processes that retard other monitored processes;

updating a database with object variable data and performance statistics in response to the occurrence of an event associated with a monitored process; and

comparing historical performance statistics from the database with current performance statistics to determine a performance trend.

2. The method of claim 1, wherein the performance trend is a trend of process response time for a given time period.

3. The method of claim 1, wherein the performance statistics include a list of processes having the longest sum in duration.

4. The method of claim 1, further comprising:

counting how many other monitored processes have been slowed down by the oldest monitored process; and

checking to determine whether to trigger the alarm for the oldest monitored process.

5. A non-transitory machine-readable storage medium storing a set of instructions that, when executed by a processor, causes the processor to perform operations, comprising:

extracting real-time data from a log file for at least one monitored process on each of the plurality of child servers with a corresponding log file reader at a parent server;

iteratively, reading command process information from the log files with the corresponding log file readers, each command process information contained in a single line, and determining how to send the single line to the parent server until the single line indicates that a corresponding child server has shut down or until the single line cannot be read;

for each monitored process, inserting extracted real-time data into object variables of an object;

processing object variable data to generate performance statistics for the monitored processes and to determine whether to trigger an alarm, the processing including identifying how many monitored processes are running in parallel and which of the running monitored processes is the oldest to identify one or more monitored processes that retard other monitored processes;

updating a database with Object variable data and performance statistics in response to the occurrence of an event associated with a monitored process; and

comparing historical performance statistics from the database with current performance statistics to determine a performance trend.

6. The non-transitory machine-readable storage medium of claim 5, wherein the performance trend is a trend of process response time for a given time period.

7. The non-transitory machine-readable storage medium of claim 5, wherein the performance statistics include statistics indicating one or more processes, if any, that retards other processes.

8. The non-transitory machine-readable storage medium of claim 5, wherein the performance statistics include a list of processes having the longest sum in duration.

9. An apparatus, comprising:

a log reader to extant real-time data from a log file for at least one monitored process on each of a plurality of child servers;

a processor for a performance monitor, the processor configured to;

iteratively read command process information from the log files with the corresponding log reader, each command process information contained in a single line, and determine how to send the single line to a parent server until the single line indicates that a corresponding child server has shut down or until the single line cannot be read;

for each monitored process, insert extracted real-time, data into object variables of an object;

process object variable data to generate performance statistics for the monitored processes and to determine whether to trigger an alarm, the processing including identifying how many monitored processes are running in parallel and which of the running monitored processes is the oldest to identify one or more monitored processes that retard other monitored processes;

update a database with object variable data and performance statistics in response to the occurrence of an event associated with, a monitored process; and

compare historical performance statistics from the database with current performance statistics to determine a performance trend.

10. The apparatus of claim 9, wherein the performance trend is a trend of process response time for a given time period.

11. The apparatus of claim 9, wherein the performance statistics include statistics indicating one or more processes, if any, that retards other processes.

12. The apparatus of claim 9, wherein the performance statistics include a list of processes having the longest sum in duration.