Skip to main content

Exciting new changes are coming to the Knowledge Base site soon!
Starting April 4, 2023, you will notice Support-Specific categorization and improvements to the search filters on the site. In May, we will be launching a new and enhanced Site UI and Navigation. To know more, read our Knowledge Article.

NetApp Knowledge Base

General guidance for collecting data for performance analysis ONTAP 7Mode and Cdot

Views:
1,035
Visibility:
Public
Votes:
0
Category:
ontap-9
Specialty:
perf
Last Updated:

Applies to

  • ONTAP 7mode
  • Cdot

Description

This is general guidance for collecting data for performance analysis.

Data collection depends on:

  • type of the issue which we need to investigate
  • protocol used for accessing data
  • infrastructure (type of network)

In some corner cases, data must be collected at multiple point in time and during different phases of reported issue.

For example, if the issue is reported for slow backup process or execution of some scrips takes longer then usually, data must be collected during whole process of backup and also during “normal” time of backup (period in which backup was not taking “much longer”).

It is crucial to collect data for performance analysis during ongoing issue which needs to be documented by providing exact timestamp.

 There are 2 major tools for collecting data:

For systems running ONTAP version 8.3 and later: Performance archive (preferred tool) and perfstat

https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/Performance_Archives%3A_Performance_first_failure_data_collection_for_Clustered_Data_ONTAP_8.3_and_later

 

For 7mode systems: Perfstat (version7.39)

https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/FAQ%3A_Perfstat_collection_methods_omnibus

Download link: https://mysupport.netapp.com/site/tools/tool-eula/perfstat7

 

Decision about timeframe for single iteration (-t parameter), number of iterations ( -i ) and number of runs may vary based on issue and is provided by TSE working on the issue via action plan.

Parameter -t (number of minutes after which counters will be collected and again zeroed) in most of the performance troubleshooting is set to 5 (5 minutes).

 

Sign in to view the entire content of this KB article.

New to NetApp?

Learn more about our award-winning Support

Scan to view the article on your device