The SUN system consisting of Nehalem and Harpertown clusters, the M9000 SMP machine, preprocessing nodes and visualization node has the following file systems:

• NFS for home directories /export/home/username
• Luster file system for fast scratch on /export/home/username/scratch

Actual file systems mounted on SUN

For example, typing the df -h command into the SUN login node will show all the available filesystems:

@login02:~> df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda3             119G   62G   52G  55% /
udev                  7.9G  900K  7.9G   1% /dev
/dev/sda1             130M   25M   98M  21% /boot
172.17.203.15:/mnt/gridware
                    1.6T  457G  1.1T  30% /opt/gridware
172.17.203.15:/mnt/home
                    1.9T  1.3T  442G  76% /export/home
172.17.203.50:/scratch/work
                    3.1T  2.6T  508G  84% /scratch/work
172.17.203.50:/scratch/home
                    712G  205G  508G  29% /scratch/home
172.17.195.21@o2ib0:172.17.195.20@o2ib0:/lfs02
                     72T   36T   33T  52% /lustre/SCRATCH2
172.17.195.20@o2ib0:172.17.195.21@o2ib0:/lfs03
                     72T   44T   25T  65% /lustre/SCRATCH3
172.17.195.21@o2ib0:172.17.195.20@o2ib0:/lfs04
                     72T  3.4T   65T   5% /lustre/SCRATCH4
172.17.195.20@o2ib0:172.17.195.21@o2ib0:/lfs01
                     72T   90G   68T   1% /lustre/SCRATCH1

There are two IBM systems (e1350 and Blue Gene/P cluster). In summary, the e1350 cluster consists of 160 compute nodes, 3 hotspares, 3 storage nodes, back-up node, management node and a single login node. These two systems are connected to the Storage Area Networks (running GPFS) of about 94 Terabytes.

Below is the structure of the file system in IBM Systems:

[chpcln][/root]> df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda2              68G   68G  1.6M 100% /
tmpfs                 8.4G  8.0K  8.4G   1% /dev/shm
/dev/home             2.8T  2.1T  677G  76% /CHPC/home
/dev/local            310G   33G  278G  11% /CHPC/usr/local
/dev/work              89T   78T   12T  88% /CHPC/work
[chpcln][/root]>

The GPU has its own designated login node connected to the 5 compute nodes and 6 storage nodes. GPU also has storage capacity of 12 Terabytes and its structured as follows:

[root@hn.gpu ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/md0               49G   16G   31G  35% /
tmpfs                  12G     0   12G   0% /dev/shm
/dev/gpfs              12T   55G   12T   1% /GPU
[root@hn.gpu ~]#