How to optimize a subsearch with a wide range?

[lmzheng]

I have a subsearch that I use to determine the first occurrence of the issue logged. I currently have an earliest=-4d, because the report runs every 15 minutes. If I were to extend the range to 365 days, the search would not be able to update efficiently.

Since the data remains the same before the current day, is it possible is cache this information so the search does not delay?

Thank you,
Lee

[richgalloway · Accepted solution]

This sounds like a job for Summary Index.

Run your reports every 15 minutes and save results in a summary index. Your search to find the first occurrence need only scan the summary rather than the raw data and should be much faster.

[lmzheng]

Is this option available for Splunk Cloud or only Splunk Enterprise?

[richgalloway]

Summary indexing is available in both Splunk Enterprise and Splunk Cloud.