I'm under splunk 5, and having a lot of scheduled Job, simply making heavy spath and rex jobs, to store my Data in kv, instead of xml, because otherwise searching is to slow...
At the moment, I use summmary_index, which have some drawbacks
- if something happens during search, summary_index is not complete
- I can't specify the sourcetype
It seems to me, that the new accelaration, seems to be stable and offer a more robust way to prepare Data, but I've seen that you need a | stats, timechart to use it ?
Is there a way to use acceleration anyway for splitting xml files ?
Otherwise is there a better way to do this in Splunk 6 than with summery_index ?
Using the Splunk 5+ Report Acceleration does indeed require a reporting command such as stats or timechart.
Depending on what you want to do with your data, you may want to take a look at accelerated data models and the pivot interface to build reports from such a data model (needs Splunk 6).
As for summary indexing and sourcetypes, you can in fact carry over things like sourcetype and host but will have to sacrifice license volume in return. I've detailed an approach here: http://answers.splunk.com/answers/88926/modify-_raw-collect-into-second-index-how-to-best-retain-hos...