1 |
Greetings, |
2 |
|
3 |
I'm currently employed at a site with some Xserve G5's and a smattering |
4 |
of PIII's. |
5 |
I cannot comment on High Availability Clusters, but I'll be more then |
6 |
willing to |
7 |
discuss the HPC side of clusters. |
8 |
|
9 |
Right now we primarily run OS X on the G5's, however work is in progress |
10 |
to allow |
11 |
job-submission time switching between OS X and Linux (Debian or Gentoo |
12 |
currently, |
13 |
others in the future possibly) based upon user-submitted requests. |
14 |
|
15 |
As we run a variety of operating systems, I personally prefer to compile |
16 |
the HPC- |
17 |
orientated applications from source. Anyways, I noticed a request for |
18 |
software |
19 |
recommendations earlier in this thread, so here's a list of the first |
20 |
things I |
21 |
end up installing when we build a test/development cluster, along with |
22 |
the versions I have running. |
23 |
|
24 |
Torque (2.0.0p5) |
25 |
Mpich (1.2.7) |
26 |
Mpichgm (Myrinet support, based on 1.2.6 ) |
27 |
Mpiexec (0.80) |
28 |
Atlas (3.7.11) |
29 |
HPL (To test the install mainly) |
30 |
|
31 |
We also find it nice to have server(s) providing: |
32 |
LDAP |
33 |
DHCP and related netbooting services. (We've written our own, highly |
34 |
alpha stage right now). |
35 |
NFS for home directories only. We've found numerous scalability |
36 |
problems with diskless. |
37 |
|
38 |
Of course the shameless plug for our MyPBS package is also required, |
39 |
http://sourceforge.net/projects/my-pbs/ |
40 |
|
41 |
This is just a quick list of of what I think any documentation on a HPC |
42 |
cluster needs to |
43 |
cover at minimum. I'm by no means an expert, but I would like to offer |
44 |
my help. |
45 |
|
46 |
Hanni Ali wrote: |
47 |
|
48 |
> Ok, |
49 |
> |
50 |
> I suggest we try to put the documentation together on gentoo-wiki.com |
51 |
> <http://gentoo-wiki.com> I've always found this site an excellent |
52 |
> resource. |
53 |
> |
54 |
> There are already two stubs which I feel we should build on and kyron |
55 |
> has compiled an excellent list of programs if you follow the links. |
56 |
> |
57 |
> http://gentoo-wiki.com/Index:HOWTO#Build_a_Gentoo_High_Performance_Cluster |
58 |
> |
59 |
> I suggest we start Build a Gentoo High Availability Cluster. |
60 |
> |
61 |
> http://www.gentoo.org/proj/en/cluster/ |
62 |
> |
63 |
> This is the gentoo cluster page and we only have three How To's All of |
64 |
> which have floors I've kept tabs on problems I've run into with the |
65 |
> HPC howto and distcc howto. I feel we should keep openMosix separate |
66 |
> and have a completely separate set of Howto's for that. |
67 |
> |
68 |
> My clusters are generally diskless nodes so I suggest we try to |
69 |
> incorporate this howto into the gentoo-wiki |
70 |
> |
71 |
> http://www.gentoo.org/doc/en/diskless-howto.xml |
72 |
> <http://www.gentoo.org/doc/en/diskless-howto.xml> |
73 |
> |
74 |
> Though this also has it's fair share of difficulties. |
75 |
> |
76 |
> I'm prepared to share a certain amount of my work on this. It would be |
77 |
> nice to make this documentation easily understandable for all and I'm |
78 |
> always up for people adding where they run into problems and WHY into |
79 |
> these sort of documents. |
80 |
> |
81 |
> I'm looking carefully at HA diskless nodes and ways in which to ensure |
82 |
> redundancy if the master node fails. Suggestions on this would be |
83 |
> welcomed. |
84 |
> |
85 |
> How many people would be interested in helping out with this. If |
86 |
> you've read this far it must be because it's a Friday afternoon so |
87 |
> anything can distract you! |
88 |
> |
89 |
> Cheers |
90 |
> |
91 |
> Hanni |
92 |
> |
93 |
> |
94 |
|
95 |
-- |
96 |
Justin Bronder |
97 |
University of Maine, Orono |
98 |
|
99 |
Advanced Computing Research Lab |
100 |
20 Godfrey Dr |
101 |
Orono, ME 04473 |
102 |
www.clusters.umaine.edu |
103 |
|
104 |
Mathematics Department |
105 |
425 Neville Hall |
106 |
Orono, ME 04469 |
107 |
|
108 |
|
109 |
-- |
110 |
gentoo-cluster@g.o mailing list |