IBM-cluster overview

General description

Acquired: April 2002 (the first 32 CPU Regatta: Fenris) and January 2003.
Vendor: IBM
System: Fenris & Sleipner IBM RS6000 - pSeries 690
Hugin & Munin: IBM RS6000 - pSeries 655
Processors: Fenris & Sleipner: 64 IBM Power4 1.3 GHz
and Hugin & Munin: 16 IBM Power4 1.1 GHz
Peak performance for the total system: 403 GFlops
L1 cache: 32 KB/64 KB for data/instructions per CPU
L2 cache: 1.5 MB shared by the two CPUs. On the chip.
L3 cache: 32 MB shared by the two CPUs.
Memory: Fenris and Sleipner: 128 GB each
Hugin & Munin: 32 GB each.
Disksystem: ca. 3.2 TB shared in a SAN based on IBM's FAStT700 RAID system
and in total ca. 3 TB local diskstorage on Fenris & Sleipner
The names Fenris, Sleipner, Hugin, Munin: from the Nordic Mythologi:
Fenris is a very strong but dangerous wulf in Asgaard. Sleipner is Odin's 8-legged horse.
Hugin and Munin is his ravens. Their names means thought and memory.

System configuration

Domainname: sleipner.cscaa.dk TCP/IP adr.: 130.225.246.21
Access: Only via SSH prot. v. 2, only for validated hosts.
Contact staff (at) cscaa.dk to get your host validated.
OS: AIX 5.1D
Batchsystem: LoadLeveler
Userfilesystem: 2.1 TB userfilesystem shared between all the nodes.
Scratchfilesystems: Fenris: 1 TB, Sleipner: 2 TB, Hugin & Munin: ca. 100 GB each.
Max. filesize of scratchfiles: 64 GB.
Scratch: Only available for running jobs.
Archive: For long-time storing of data. Unlimited size (in principle).
Accounting: Jobs will be accounted according to the wall clock time used.
1 BU (Bonus Unit) equals 90 minuttes.
Backup: An incremental backup of the userfilesystem is performed every night.
The scratchfilesystems are never backed up.


		ABOUT-DCSC	DCSC/SDU	DCSC/AU	DCSC/AAU	DCSC/DTU	DCSC/KU

	+Open all -Close all About DCSC /AU Home Information About DCSC /AU Organisation News Current News News Archive Resources Grendel cluster Grendel GPU cluster Huge IBM p6-cluster Freke SGI Altix Sleipner IBM p4-cluster Karlsen SGI Origin 2000 The Archive Grid Computing Statistics Utilization Graphs System Load Support How to Get Help How to Get a Grant How to Get an Account Utillities Links Contact DCSC /AU Contact DCSC Central Office Other DCSC Centers
			Overview Hardware Software Batchjobs Hints IBM-cluster overview General description Acquired: April 2002 (the first 32 CPU Regatta: Fenris) and January 2003. Vendor: IBM System: Fenris & Sleipner IBM RS6000 - pSeries 690 Hugin & Munin: IBM RS6000 - pSeries 655 Processors: Fenris & Sleipner: 64 IBM Power4 1.3 GHz and Hugin & Munin: 16 IBM Power4 1.1 GHz Peak performance for the total system: 403 GFlops L1 cache: 32 KB/64 KB for data/instructions per CPU L2 cache: 1.5 MB shared by the two CPUs. On the chip. L3 cache: 32 MB shared by the two CPUs. Memory: Fenris and Sleipner: 128 GB each Hugin & Munin: 32 GB each. Disksystem: ca. 3.2 TB shared in a SAN based on IBM's FAStT700 RAID system and in total ca. 3 TB local diskstorage on Fenris & Sleipner The names Fenris, Sleipner, Hugin, Munin: from the Nordic Mythologi: Fenris is a very strong but dangerous wulf in Asgaard. Sleipner is Odin's 8-legged horse. Hugin and Munin is his ravens. Their names means thought and memory. System configuration Domainname: sleipner.cscaa.dk TCP/IP adr.: 130.225.246.21 Access: Only via SSH prot. v. 2, only for validated hosts. Contact staff (at) cscaa.dk to get your host validated. OS: AIX 5.1D Batchsystem: LoadLeveler Userfilesystem: 2.1 TB userfilesystem shared between all the nodes. Scratchfilesystems: Fenris: 1 TB, Sleipner: 2 TB, Hugin & Munin: ca. 100 GB each. Max. filesize of scratchfiles: 64 GB. Scratch: Only available for running jobs. Archive: For long-time storing of data. Unlimited size (in principle). Accounting: Jobs will be accounted according to the wall clock time used. 1 BU (Bonus Unit) equals 90 minuttes. Backup: An incremental backup of the userfilesystem is performed every night. The scratchfilesystems are never backed up.