1 |
On 1/31/07, Brian D. Ropers-Huilman <bropers@×××××××.edu> wrote: |
2 |
> |
3 |
> Your comment on needing to "store *a lot* of data (like 50TB max)" is very |
4 |
> relative. For sites like ours and many others, *a lot* of data means 100's |
5 |
> of PB. :) |
6 |
> |
7 |
> As to your questions RE: GlusterFS, this announcement has also recently |
8 |
> made |
9 |
> the rounds on the beowulf mailing list though there has not been too much |
10 |
> conversation about it. Being new, nobody has tried it as of yet, as such |
11 |
> there are not many comparative studies (if any) out there. |
12 |
|
13 |
|
14 |
My thinking exactly, it's very easy to say it does a lot of things. |
15 |
|
16 |
What they claim to be able to do is quite nice. But then again, many vendors |
17 |
> and other FOSS cluster file systems have made such claims in the past. |
18 |
> Only |
19 |
> time will tell. |
20 |
> |
21 |
> The only real thing I can offer is that I will be watching this very |
22 |
> closely |
23 |
> which might lend some credence to Gluster as a potential solution. |
24 |
|
25 |
|
26 |
I'll see if I can test it here at my work somewhen between now and the next |
27 |
3 weeks. I would like to see how failsafe it really is. As soon as I have |
28 |
the results I'll post it here. |
29 |
|
30 |
-- |
31 |
> Brian D. Ropers-Huilman, Director |
32 |
> Systems Administration and Technical Operations <bropers@×××××××.edu> |
33 |
> Supercomputing Institute for Digital Simulation and Advanced Computation |
34 |
> 599 Walter Library +1 612-626-5948 (V) |
35 |
> 117 Pleasant Street S.E. +1 612-624-8861 (F) |
36 |
> University of Minnesota Twin Cities Campus |
37 |
> Minneapolis, MN 55455-0255 http://www.msi.umn.edu/ |
38 |
> -- |
39 |
> gentoo-cluster@g.o mailing list |
40 |
> |
41 |
> |