Large scale Apache log file analysis

3 points by bdwalter ↗ HN
Does anyone know of any good commercial products to analyze about 500 million to 1 billion rows (1 week) of apache access logs?

Looking for fairly simple analysis. Something similar to what weblog expert does, but on a larger scale.

Are there any commercial tools out to do this or are we stuck writing something for this?

I know we could use hive/hadoop/cassandra etc and build something. We are about to roll out another product that uses cassandra as a backend, but this is simply to get an idea of some of our basic web stats in a simple report. This is not part of our core product, so I am just looking to see if there is an easy way to do something here without a bunch of development.

2 comments

[ 3.1 ms ] story [ 12.1 ms ] thread
Splunk is very popular, although I think it's software that you have to host. I have no idea how fast it is, but with a recent-ish machine I wonder if you could have it process stuff in a weekly batch with minimum of frustration. They have a free trial that might be worth it to you -- www.splunk.com