thanks for the tips so far.  I should have been a bit more specific.  It's 
Saturday toady and I'm doing this off the top of my head so I might be off 
by a bit but as I recall in Splunk right now we have the equivalent to 11 
indexes - the biggest one runs 4Gb a day, all together they run 1.2Tb a 
day.  We retain the data for 90 days.  We have 12 machines indexing the 
data in EC2 (m2.4xlarge) and although it works fine it is too slow (users 
complain about report speed).  

If EC works the money I can save from not renewing my spunk license could 
easily double the number of servers and upgrade them to the i class (SSD 
storage with big ram) *and* send the team to Europe for a couple of weeks 
(although the trip to Europe is not my decision).  

I will look for the Goldman Sachs talk.  My plan after reading the ES 
website is to leave Splunk alone, fork the data for one index to a new ES 
cluster and Splunk then make the comparisons.  My only issue is if I go 
with the i instances (with SSD's) it's not a fair comparison for 
benchmarking.  That may not be a big deal for me but I'd love to see the 
Apples to Apples numbers.

Frank

-- 
You received this message because you are subscribed to the Google Groups 
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/6dafa0bb-3616-476e-9409-0fed8b47dd86%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to