JBOD as talked about here http://www.datastax.com/wp-content/uploads/2012/08/C2012-StateofCassandra-JonathanEllis.pdf and defined by disk_failure_policy

So that when you have very large nodes disk failed does not require a full replacement. But if you are using a high level raid guess that's not necessary. 

Cheers

-----------------
Aaron Morton
Freelance Cassandra Consultant
New Zealand

@aaronmorton

On 2/04/2013, at 6:35 PM, "Hiller, Dean" <Dean.Hiller@nrel.gov> wrote:

Oh, JBOD, not JBOB.

no, we were using RAID 5 and RAID 6 from what I understand.  I am trying to get a test run with just one disk to make sure the test is correct as one disk should have much less performance than 20 in the case of random access.  In sequential, I think performance would be the same(ie. Both would be 250MB/sec in throughput is my guess)
Thanks,
Dean

From: <Hiller>, Nrel <Dean.Hiller@nrel.gov<mailto:Dean.Hiller@nrel.gov>>
Date: Tuesday, April 2, 2013 6:40 AM
To: "user@cassandra.apache.org<mailto:user@cassandra.apache.org>" <user@cassandra.apache.org<mailto:user@cassandra.apache.org>>
Subject: Re: how to test our transfer speeds

Is 1.2 JBOB and april fools joke?  Heh, seriously though, I have no idea what you are talking about there.  I am trying to get raw disk performance with no cassandra involved before involving cassandra…..which is the next step.

Thanks,
Dean

From: aaron morton <aaron@thelastpickle.com<mailto:aaron@thelastpickle.com>>
Reply-To: "user@cassandra.apache.org<mailto:user@cassandra.apache.org>" <user@cassandra.apache.org<mailto:user@cassandra.apache.org>>
Date: Monday, April 1, 2013 11:01 PM
To: "user@cassandra.apache.org<mailto:user@cassandra.apache.org>" <user@cassandra.apache.org<mailto:user@cassandra.apache.org>>
Subject: Re: how to test our transfer speeds

If not, maybe I just generate the same 1,000,000 files on each machine, then randomly delete 1/2 the files and stream them from the other machine as writing those files would all be in random locations again forcing a much worse measurement of MB/sec I would think.
Not sure I understand the question. But you could just scrub the data off a node and rebuild it.

Note that streaming is throttled, and it will also generate compaction.

He has twenty 1T drives on each machine and I think he also tried with one 1T drive seeing the same performance which makes sense if writing sequentially
Are you using the 1.2 JBOB configuration?

Cheers

-----------------
Aaron Morton
Freelance Cassandra Consultant
New Zealand

@aaronmorton
http://www.thelastpickle.com

On 1/04/2013, at 11:01 PM, "Hiller, Dean" <Dean.Hiller@nrel.gov<mailto:Dean.Hiller@nrel.gov>> wrote:

(we plan on running similar performance tests on cassandra but wanted to understand the raw foot print first)…..

Someone in ops was doing a test transferring 1T of data from one node to another.  I had a huge concern I emailed him that this could end up being a completely sequential write not testing random access speeds.  He has twenty 1T drives on each machine and I think he also tried with one 1T drive seeing the same performance which makes sense if writing sequentially.  Does anyone know of something that could generate a random access pattern such that we could time that?  Right now, he was measuring 253MB / second from the time it took and the 1T of data.  I would like to find the much worse case of course.

If not, maybe I just generate the same 1,000,000 files on each machine, then randomly delete 1/2 the files and stream them from the other machine as writing those files would all be in random locations again forcing a much worse measurement of MB/sec I would think.

Thoughts?

Thanks,
Dean