Home
Systemutvecklare med inriktning mot storskalig lagring
Halvan needs to be restarted due to a memory overflow
SNIC Cloud Computing Course at UPPMAX, May 16
The Grad cluster
grad cluster
Grad provides 512 CPU cores in the form of 64 dual CPU (Intel 2.66GHz
quad core, E5430) compute nodes with 16 GByte RAM per node.
By using "twin nodes" with dual nodes in each 1u rack mounted unit the system has been made very dense and easily fits into a single rack.
Grad serves as a compute cluster in the national Swegrid computational grid, providing computing power to national and international large scale computing projects.
Grad also include storage elements for Swegrid providing 140TByte of dCache disk pools.
By using "twin nodes" with dual nodes in each 1u rack mounted unit the system has been made very dense and easily fits into a single rack.
Grad serves as a compute cluster in the national Swegrid computational grid, providing computing power to national and international large scale computing projects.
Grad also include storage elements for Swegrid providing 140TByte of dCache disk pools.
The system was delivered in spring 2008 and started full scale operatiom in september.
There is a rudimentary Grad user guide available for those interested in more details or learning how to use the system.
See the MPI and OpenMP tutorial for information on how to compile and run parallel programs on Uppmax systems.
Technical Summary
- 64 Compute Nodes with 512 CPU cores
- 3 access Nodes with 8 CPU cores
- 16 Gigabyte RAM per node
- 18 Terabyte disk in 2 NFS servers for home directories and grid sessions
- 140 Terabyte disk in 8 nodes of dCache pools
- switched Gigabit Ethernet for all nodes
