I have not looked at Lustre, as I have heard many negative things about it
(including Oracle ownership). The only business using Lustre where I know
the admins has had a lot of trouble with it. No redundancy.
I know some Lustre admins that indeed have the far away stare similar to
people that
On Sun, 18 May 2014, Ted Miller wrote:
How recently have you looked at Gluster? It has seen some significant
progress, though small files are still its weakest area. I believe that
some use-cases have found that NFS access is faster for small files.
I last looked at Gluster about two months
On Sun, 18 May 2014, Les Mikesell wrote:
Do you really need filesystem semantics or would ceph's object store work?
Yes, I really need file system semantics; I am storing home directories.
Steve
___
CentOS mailing list
CentOS@centos.org
We were using glusterfs for shared home directories and it was really slow.
We're using an NFS shared and it's working much faster.
Mark
On May 18, 2014, at 21:35, Ted Miller tedli...@sbcglobal.net wrote:
On 05/18/2014 11:47 AM, Steve Thompson wrote:
MooseFS and GlusterFS have both been
On Mon, May 19, 2014 at 6:35 AM, Steve Thompson s...@vgersoft.com wrote:
On Sun, 18 May 2014, Les Mikesell wrote:
Do you really need filesystem semantics or would ceph's object store work?
Yes, I really need file system semantics; I am storing home directories.
In that case, wouldn't it be
On Sun, 18 May 2014, Andrew Holway wrote:
Have you looked at parallel filesystems such as Lustre and fhgfs?
I have not looked at Lustre, as I have heard many negative things about it
(including Oracle ownership). The only business using Lustre where I know
the admins has had a lot of trouble
On 05/18/2014 11:47 AM, Steve Thompson wrote:
MooseFS and GlusterFS have both been evaluated, and were too slow. In the
case of GlusterFS, wy too slow.
How recently have you looked at Gluster? It has seen some significant
progress, though small files are still its weakest area. I believe
On Sun, May 18, 2014 at 10:47 AM, Steve Thompson s...@vgersoft.com wrote:
On Sun, 18 May 2014, Andrew Holway wrote:
MooseFS and GlusterFS have both been evaluated, and were too slow. In the
case of GlusterFS, wy too slow.
Do you really need filesystem semantics or would ceph's object
This idea is intruiging...
Suppose one has a set of file servers called A, B, C, D, and so forth, all
running CentOS 6.5 64-bit, all being interconnected with 10GbE. These file
servers can be divided into identical pairs, so A is the same
configuration (diks, processors, etc) as B, C the same
On Sat, May 17, 2014 at 10:30 AM, Steve Thompson s...@vgersoft.com wrote:
This idea is intruiging...
Suppose one has a set of file servers called A, B, C, D, and so forth, all
running CentOS 6.5 64-bit, all being interconnected with 10GbE. These file
servers can be divided into identical
On Sat, 17 May 2014, SilverTip257 wrote:
Sounds like you might be reinventing the wheel.
I think not; see below.
DRBD [0] does what it sounds like you're trying to accomplish [1].
Especially since you have two nodes A+B or C+D that are RAIDed over iSCSI.
It's rather painless to set up
How about glusterfs?
17.5.2014 20.01 kirjoitti Steve Thompson s...@vgersoft.com:
On Sat, 17 May 2014, SilverTip257 wrote:
Sounds like you might be reinventing the wheel.
I think not; see below.
DRBD [0] does what it sounds like you're trying to accomplish [1].
Especially since you have
On Sat, 17 May 2014, Eero Volotinen wrote:
How about glusterfs?
I have tried glusterfs; the large file performance is reasonable, but
the small file performance is too low to be useable.
Steve
___
CentOS mailing list
CentOS@centos.org
On Sat, May 17, 2014 at 1:00 PM, Steve Thompson s...@vgersoft.com wrote:
On Sat, 17 May 2014, SilverTip257 wrote:
Sounds like you might be reinventing the wheel.
I think not; see below.
DRBD [0] does what it sounds like you're trying to accomplish [1].
Especially since you have two
On 17.05.2014 19:00, Steve Thompson wrote:
On Sat, 17 May 2014, SilverTip257 wrote:
Sounds like you might be reinventing the wheel.
I think not; see below.
DRBD [0] does what it sounds like you're trying to accomplish [1].
Especially since you have two nodes A+B or C+D that are RAIDed
On Sun, 18 May 2014, Dennis Jacobfeuerborn wrote:
Why specifically do you care about that? Both with your solution and the
DRBD one the clients only see a NFS endpoint so what does it matter that
this endpoint is placed on one of the storage systems?
The whole point of the exercise is to end
Have you looked at parallel filesystems such as Lustre and fhgfs?
On 18 May 2014 01:14, Steve Thompson s...@vgersoft.com wrote:
On Sun, 18 May 2014, Dennis Jacobfeuerborn wrote:
Why specifically do you care about that? Both with your solution and the
DRBD one the clients only see a NFS
17 matches
Mail list logo