This unit will be (eventually) the replacement for our older central server at our new space (woo-hoo!!!!). Right now, taking to the test track as it were.
Simple machine: 16 GB ram, 4 cores, 7.5 TB of raw storage. In a deskside case. Works well for offices. This configuration would be right about $5900 list. RAID6 with one hot spare would drop it to 6TB for storage. Carving out 2 drives for OS (as I did) would bring it down to 5TB.
root@crunch:~# df -h /big
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 4.6T 36G 4.6T 1% /big
root@crunch:~# df -m /big
Filesystem 1M-blocks Used Available Use% Mounted on
/dev/sdb1 4768242 36783 4731459 1% /big
root@crunch:~# df /big
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sdb1 4882679680 37665724 4845013956 1% /big
Well, with rounding for all the conversions, call it 4.6TB usable.
Bonnie and iozone numbers put it 500-600 MB/s sustained. Obviously we have some tuning to do on this (using a new kernel, so we need to adjust to it).
root@crunch:~# uname -a
Linux crunch.scalableinformatics.com 2.6.24-16-server #1 SMP Thu Apr 10 13:15:38 UTC 2008 x86_64 GNU/Linux
Previous tests on 2.6.24 have shown it to be quite performance challenged, so we will likely work with a 2.6.25 build soon.
We are running mpiblast 1.5.0 pio on the unit to do some testing for a customer. At the same time, for laughs, I ran a quick dd command with direct IO. So we aren’t getting all the IO time, processor time, and there is contention (4 cores of 4 in use by mpiblast with IO in flight). Still, results aren’t shabby.
root@crunch:~# dd oflag=direct if=/dev/zero of=/big/file bs=1M count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 15.3083 s, 561 MB/s