<div dir="ltr">I am using 3 node cluster , with 3 copies of disk image. So, each node have a full copy of disk image. <br></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Jun 23, 2015 at 10:37 AM, Vasiliy Tolstov <span dir="ltr"><<a href="mailto:v.tolstov@selfip.ru" target="_blank">v.tolstov@selfip.ru</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">2015-06-23 1:40 GMT+03:00 ankit bhardwaj <<a href="mailto:ankitbhrdwj27@gmail.com">ankitbhrdwj27@gmail.com</a>>:<br>
> I am new to Sheepdog. Though I have gained enough knowledge to deploy and<br>
> use sheepdog for qemu-kvm VMs. I tried to benchmark Sheepdog and found that:<br>
><br>
> 1. Sheepdog's Maximum Read/Write Speed is around 70-80% compared to local<br>
> storage. That too when the condition are in favour of sheepdog. Example:<br>
> Full copy of data on same node and Reading or writing sequential Data etc.<br>
><br>
> 2. When using VM disk image to boot VM from Sheepdog storage, some extra<br>
> system activities increase the load on the CPU from 30-40%.<br>
><br>
> Can you please tell me,:<br>
> Why the speed is slow, even if all the data is on the same machine ?<br>
> What activity in sheepdog is eating up my processor cycles ?<br>
><br>
><br>
> Sorry for the loosely described experiments and results.<br>
<br>
<br>
</div></div>I think we need command line for running sheepdog and how you format<br>
you cluster (erasure coded or replicas) ?<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Vasiliy Tolstov,<br>
e-mail: <a href="mailto:v.tolstov@selfip.ru">v.tolstov@selfip.ru</a><br>
</font></span></blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature">Ankit Bhardwaj</div>
</div>