Re: Initial Phase of Deploying Openstack

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



You guys are great. Thank you very much for the assistance!



Regards,

Christopher Coulson
-------------------------------------------------------
Systems Administrator
CPI Group, Inc.
3719 Corporex Park Drive, Suite #50
Tampa, FL 33619
Phone: 813.254.6112 (ext. 635)
Fax:   813.514.0637
Email: chrisc@xxxxxxxxxxxxxxx
-------------------------------------------------------


On Tue, Apr 30, 2013 at 4:15 PM, Gregory Farnum <greg@xxxxxxxxxxx> wrote:
[ Please keep discussions on the list — thanks! :) ]

On Tue, Apr 30, 2013 at 1:09 PM, Chris Coulson <chrisc@xxxxxxxxxxxxxxx> wrote:
> First, thank you for your reply. I guess I wasn't specific enough-- my
> question really should have been: when planning to use Ceph, should there be
> any specific hardware considerations for a consolidated deployment on a
> single node? We're interested in taking advantage of both the object-based
> and block storage offered by Ceph, and I'm not sure if we could simply pick
> up a basic DAS/NAS storage server, install Ceph, and be good to go, or if
> we're missing something.

Ah, yeah. Just pick a server that can hold a bunch of drives. We've
seen that some of them have issues with eg oversubscribed SAS/SATA
expanders so you'll want to check their basic disk capability, and you
want enough compute power to handle each drive (we recommend 1GHz of
CPU and 1GB of RAM per OSD/disk), but within those constraints you can
go wild.
Like John said, you probably want more than one server for a
production deployment (just in terms of reliability), but in terms of
the software you can configure everything to work that way.
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com

>
>
>
> Regards,
>
> Christopher Coulson
> -------------------------------------------------------
> Systems Administrator
> CPI Group, Inc.
> 3719 Corporex Park Drive, Suite #50
> Tampa, FL 33619
> Phone: 813.254.6112 (ext. 635)
> Fax:   813.514.0637
> Email: chrisc@xxxxxxxxxxxxxxx
> -------------------------------------------------------
>
>
> On Tue, Apr 30, 2013 at 4:03 PM, Gregory Farnum <greg@xxxxxxxxxxx> wrote:
>>
>> On Tue, Apr 30, 2013 at 12:35 PM, Chris Coulson <chrisc@xxxxxxxxxxxxxxx>
>> wrote:
>> > Hopefully I'm performing the correct task to ask a quick,
>> > facepalm-inducing
>> > question:
>> >
>> >
>> > After attending the Openstack Summit in Portland, my company is planning
>> > to
>> > implement our own private cloud and are beginning to develop ideas
>> > regarding
>> > its architecture. Skipping extraneous information: I'm curious to know
>> > if
>> > it's possible to deploy Ceph (we like the idea of combining block and
>> > object-based storage instead of using 'the other guys' separately) on a
>> > SINGLE storage server, and if so-- how would you recommend it be done?
>> > DAS?
>> > NAS? Obviously this is not ideal for failover or redundancy, but for our
>> > initial configuration, we will likely be going this route.
>>
>> I'm not quite sure what you're asking about here. It's perfectly
>> possible to deploy Ceph on a single node; just run an OSD daemon per
>> drive, and a monitor daemon on a drive. You'd have to connect to it
>> through the interfaces you're interested in.
>> -Greg
>> Software Engineer #42 @ http://inktank.com | http://ceph.com
>
>

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux