benchmark results on my $75/month 5 node cluster :)

Alex Rice alex at mindlube.com
Thu Oct 3 11:43:14 EDT 2013


Thanks for the link, James!

Windows Azure has an odd SLA- in fact if you read the fine print they
dont even offer SLA for individual VMs. They reserve the right to take
down individual VMs if they need to do an upgrades of the Host OS. But
they offer  'availability zones' so you can make sure only one VM gets
rebooted at a time. Anyways- Riaks failover-free distributed design
fits this very scenario well.

I've also tried Amazon, Cloudsigma, and Softlayer, but Azure is my
actually favorite. It's not the cheapest but it just seems to work for
me.

On Thu, Oct 3, 2013 at 8:46 AM, James Martin <jmartin at basho.com> wrote:
> Alex,
>
> Nice work benchmarking!  One thing to keep in mind (and you probably
> are well aware of this), particularly when benchmarking on cloud
> platforms, is that there are a lot of factors that affect performance
> outside  -- other tenants, blackbox hardware upgrades, changes in SLAs
> -- just to name a few.  What you may see today in your benchmarks may
> not be what you see tomorrow, or even in the next hour.  Some
> providers do offer performance SLAs and/or guaranteed IOPs, and this
> can help mitigate the problem, but doesn't necessarily remove it.
>
> You might find this article interesting (although a bit dated):
> http://blog.scalyr.com/2012/10/16/a-systematic-look-at-ec2-io/




More information about the riak-users mailing list