Tracing performance issues


(courtenay) #1

Hey!

What is your favorite way to closely inspect query performance?

We recently upgraded from 0.19.2 to 0.90.5, and moved off AWS onto
dedicated hardware. The new cluster runs at a much lower load, and is on
SSDs.

You'd think that performance would increase vastly, but instead it dropped

  • our queries are now awful, are around 400ms. Is there a way of seeing
    where the breakdown is? For example, mysql has EXPLAIN... which can
    indicate index issues.

-- Courtenay

--
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 elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/db27aa7b-7907-4e05-82c9-236992be08d7%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.


(Mark Walkom) #2

You never mentioned what OS, but assuming linux you might want to look at
this - http://dev.nuodb.com/techblog/tuning-linux-io-scheduler-ssds

Regards,
Mark Walkom

Infrastructure Engineer
Campaign Monitor
email: markw@campaignmonitor.com
web: www.campaignmonitor.com

On 10 December 2013 07:27, courtenay court3nay@gmail.com wrote:

Hey!

What is your favorite way to closely inspect query performance?

We recently upgraded from 0.19.2 to 0.90.5, and moved off AWS onto
dedicated hardware. The new cluster runs at a much lower load, and is on
SSDs.

You'd think that performance would increase vastly, but instead it dropped

  • our queries are now awful, are around 400ms. Is there a way of seeing
    where the breakdown is? For example, mysql has EXPLAIN... which can
    indicate index issues.

-- Courtenay

--
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 elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/db27aa7b-7907-4e05-82c9-236992be08d7%40googlegroups.com
.
For more options, visit https://groups.google.com/groups/opt_out.

--
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 elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAEM624aPa-TPA6NuQOuJ2eMWjt9r6HKoEcuqxy5SeRAc5j26vA%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.


(Jörg Prante) #3

Can you give more info about CPU, RAM, OS, JVM? Maybe network? How does
your queries look like?

You shoud use monitoring tools to watch your query workload and if your
resources can deal with it.

Jörg

--
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 elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAKdsXoGJiGUGfXxTW9bTYk4KO7F323EFD%2BxLerRR04XLSPH2jQ%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.


(system) #4