[Gluster-users] GlusterFS & XenServer Baremetal
Alexandre Emeriau
cloudbeans at gmail.com
Sat Jul 18 01:51:14 UTC 2009
So this is not doable :) in regards to :
* more than the throughput of a single server, if not in overall speed, then
in width.
* being able to add more bricks in and have the expectation of increased
storage capacity and throughput.
No hope or ideas for getting a better situation ?
If not, what scenario did you switch for ?
On Sat, Jul 18, 2009 at 3:21 AM, Nathan Stratton <nathan at robotics.net>wrote:
> On Sat, 18 Jul 2009, Alexandre Emeriau wrote:
>
> Hello,
>>
>> What is for you the best GlusterFS scenario in using XenServer (i'm not
>> talking about Xen on a linux but XenServer baremetal) for a web farm
>> (Apache-Tomcat) ? I were thinking of using ZFS as the filesystem for the
>> different nodes.
>>
>> The objectives/needs :
>>
>> * A storage cluster with the capacity equal to at least 1 node(assuming
>> all
>> nodes are the same).
>> * being able to lose/take down any one brick in the cluster at any time
>> without a loss of data.
>> * more than the throughput of a single server, if not in overall speed,
>> then
>> in width.
>> * being able to add more bricks in and have the expectation of increased
>> storage capacity and throughput.
>> * present the storage as a single entity
>>
>
> This is all doable, the only downside is that last time I tested even with
> raw RAID 6 IO of 400 meg, I was only able to get 25 MB/s with
> --disable-direct-io-mode.
>
> <>
>>
> Nathan Stratton CTO, BlinkMind, Inc.
> nathan at robotics.net nathan at blinkmind.com
> http://www.robotics.net http://www.blinkmind.com
>
More information about the Gluster-users
mailing list