I also saw somewhere (may have been twitter) that the reads they benchmark against for provisioned IOPS is something like 4KB or 8KB. It was something small, smaller than the page size memmapping will use anyway. 

Cheers

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

@aaronmorton
http://www.thelastpickle.com

On 14/08/2013, at 1:23 PM, Ben Bromhead <ben@instaclustr.com> wrote:

 http://www.mail-archive.com/user@cassandra.apache.org/msg11022.html sums it up pretty well. Optimised images and provisioned IOPS may help, but whatever way you spin it your reads and writes are still going out on the network somewhere.

EBS is like a giant SAN which will drop out at any second, take almost everything in your region down with it whilst simultaneously opening up a gate to hell that lets all sorts of unimaginable horrors into the world. 

Ok maybe not that bad, but network issues between ebs and your instances is painful. Whereas network issues with a single AZ can be dealt with in the course of normal cluster operations.

On a slight tangent, have a read of http://thelastpickle.com/2011/06/13/Down-For-Me/ which does an awesome job of explaining what will happen to your quorum reads and writes when a AWS AZ goes down (and you use ephemeral storage).

Cheers

Ben Bromhead
Instaclustr | www.instaclustr.com | @instaclustr | +61 415 936 359


On 14/08/2013, at 10:42 AM, Jon Haddad <jon@jonhaddad.com> wrote:

I strongly recommend against EBS, even with optimized & ebs provisioned.  The throughput you'll get from local drives is significantly better than what you'll get with EBS (even 4K iops provisioned)

On Aug 13, 2013, at 2:10 PM, Rahul Gupta <rgupta@dekaresearch.com> wrote:

I am working on requirement to host multi tenant Cassandra cluster (or set of clusters) on Amazon EC2 (AWS).
 
With everything else sorted out, I have below question where I am looking for recommendations:
 
Does Amazonís recent support of EBS optimized images changes whole discussion around EBS vs. ephemeral drives and image size?
 
         Option 1: reserved m1.xlarge (4x420GB drives) is $0.187/hr
         Option 2: reserved m1.large EBS-optimized  is $0.119/hr (~$50/month less than m1.xlarge, but $168/month for 4x420 standard EBS volumes): costs $120/month more, but additional recovery options
 
Given Cassandra is designed to survive failures, combining replication factor 3 and backing-up to S3, I think should be enough for back up.
 
Please advise.
 
Thanks,
Rahul Gupta
DEKA Research & Development
340 Commercial St  Manchester, NH  03101
P: 603.666.3908 extn. 6504 | C: 603.718.9676
 

This e-mail and the information, including any attachments, it contains are intended to be a confidential communication only to the person or entity to whom it is addressed and may contain information that is privileged. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please immediately notify the sender and destroy the original message.

 


This e-mail and the information, including any attachments, it contains are intended to be a confidential communication only to the person or entity to whom it is addressed and may contain information that is privileged. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please immediately notify the sender and destroy the original message.

Thank you.

Please consider the environment before printing this email.