Data model split row 1M limit

[hylam]

I am calculating the average number of page views per session in a web server log file. I have accelerated the data model. The 1M limit is smaller than expected for a big data map reduce technology. How does it scale up or scale out? I will consider sampling if this doesn't scale. Thx.

sourcetype=access_combined |stats count as cnt by session_id | stats avg(cnt)
(Note: I did not test this search)

[martin_mueller · Accepted solution]

A zero usually tells Splunk to not apply any limit.

[martin_mueller]

Data models scale with the regular horizontal scaling of Splunk, by adding more indexers.

[hylam]

How does it scale up or scale out?