Computational facilities
Beowulf clusters looking for new frontiers.
- Hexa
This Beowulf cluster contains 36 nodes with dual processor, hexa core 2.67 GHx EM64T Xeon processors (5650) all with Hyper-Threading enabled. In total there are 432 cores for computing available. The nodes are connected by HP Procurve 2848 gigabit switch.
- Read more
- Gainestown
This Beowulf cluster contains 28 nodes with dual processor, quad core 2.40 GHx EM64T Xeon processors (5530) and 6 nodes with dual processor, quad core 2.53 GHx EM64T Xeon processors (5630) all with Hyper-Threading enabled. In total there are 272 cores for computing available.
- Read more
- Nehalem
This Beowulf cluster contains 16 nodes with dual processor, quad core 2.40 GHx EM64T Xeon processors (5530) and 12 nodes with dual processor, quad core 2.53 GHx EM64T Xeon processors (5630) all with Hyper-Threading enabled. In total there are 224 cores for computing available.
- Read more
- Harpertown
-
This Beowulf cluster contains 32 nodes with dual processor, quad core 2.67 GHx EM64T Xeon processors (5430). In total there are 256 cores for computing available. The nodes are connected by HP Procurve 2848 gigabit switch. Each node has 16Gb DDRII 1333 MHz RAM and two 150 GB 10Krpm SATA harddisks available. On the head node a total storage capacity of 1.5 TB, using software RAID-5 over seven 300 Gb SATA disks (one hot spare), is available and with the system disks running from a software based RAID-1 configuration.
This system runs on CentOS 5.6.
- Read more
- Woodcrest
-
This Beowulf contains 20 nodes with dual processor, dual core 2.67 GHx EM64T Xeon processors (5150), 8 nodes with dual processor, quad core 2.33 GHz EM64T Xeon processors (5345) and 6 nodes with dual processor, quad core 2.66 GHz EM64T Xeon processors (5430). In total there are 192 cores for computing available. The nodes are connected by HP Procurve 2848 gigabit switch. Each dual core node (5150) has 8Gb DDRII 1333 MHz RAM. The quadcore nodes (5345 and 5430) have 16Gb RAM available. All nodes have two 150 GB 10Krpm SATA harddisks available. On the head node a total storage capacity of 1.8 TB, using software RAID-6 over four 1 TB SATA disks, is available and with the system disks running from a software based RAID-1 configuration. This system runs on Scientific Linux 4.3.
- Read more
- Server room
-
Currently all our clusters and servers are located in the server room in the Gorlaeus.
- SARA Huygens
-
We frequently make use of the IBM Power6 national supercomputer called Huygens. This newly installed national supercomputer is located at the SARA institute in Amsterdam.
- Read more
- BOINC desktop grid
-
Our group also has access to a BOINC based desktop grid, which at the moment, consists of more than 68k computers. On this grid students and reachers can run classical trajectory calculations through a personalised queuing system we developed here in Leiden. More information on this grid can be found at http://boinc.gorlaeus.net. The current state of this grid can be viewed at: http://boinc.gorlaeus.net/totals.php.
- Read more
- Leiden Grid Infrastructure
-
Currently we have connected all these computational resources in a computer grid. For this we developped a grid middleware with the support of NWO-NCF. The report of that project and the LGI software can be found here.
- Read more


