New Submit Machine
I cooked a new "cinnnamon" spice submit machine today. It's got a RAID made from 280 GB disks resulting in 699 GB raw and 698 GB of usable disk space.
Monitoring PhEDEx
We're monitoring PhEDEx now. The result, in Cnagios is... well... too wide to fit this blog. Trust me. It's in there. Try the web interface.
Storage Server IO Graphs
About 50 lines of perl later and now the dCache server network IO graphs semi-automatically maintained...
http://noc.hep.wisc.edu/ad-hoc-graphs/StorageServerNetIO.cgi
Monitoring New GridFTP Doors
We're monitoring port 2811 on the new GridFTP doors (whatever they are?!)
Monitoring RAIDs on s5nXX
We're monitoring their utilization at least. The actual
RAID status monitoring will happen really soon now.
Impressive Data Rates
I'm rather certain we set a new record for egress traffic here at the University of Wisconsin. For a few minutes we peaked at 4.2 Gbps and we sustained 3.3 Gbps for a little over 30 minutes. The end-to-end (Fermi to UW HEP) applications (PHEDeX) data rate was around 2.7 Gbps (330 MBps.)
Disks for the g7nXX Systems
We recently inherited 27 dual 3.0 GHz Xeon compute servers (which were already housed in our machine room.) In order to increase our storage space for our CMS Tier-2 storage facility, we're going to install a bunch of disks in them.
Yesterday I asked for quotes for qty 55 750 GB Seagate 7200.10 disks, but then on vendor said they've seen a 80% failure rate with that drive. So now we're looking at buying 55 500 GB Seagate drives--the NL35.2 model--which has a better MTBF. Oh, well, only another 13.5 TB instead of 20.
500 GB ATA Disks for the g8nXX Systems
Will installed them. I configured them. Together we added another 6 TB of disk space
More dCache Storage
We ordered 12 Seagate Barracuda 7200.9 ATA100 500 GB hard drives today. They'll go into the g8nXX compute nodes and add another ~6 TB of disk space to our dCache pool.
Hardware Update for DISUN Annual Report
At the request of Frank Wuerthwein, I wrote up the following blurb, fwiw:
In year one of DISUN, UW-HEP expanded it's CMS Tier-2 cluster by adding 184 CPUs (46 dual-CPU/dual-core 1.8 GHz Opterons providing around 221 kSpecInt2000.) We estimate the total performance capacity of our CMS Tier-2 dedicated computing resources at around 675 kSpecInt2000. In addition, we added 90 TB of disk space. The total upgrade costs add up to around $175,000, of which $126,000 was funded by DISUN.
