wir bieten...
Dekobild im Seitenkopf ISMLL
 

Rocks Cluster at ISMLL

ISMLL has a Rocks Cluster (Release 6.1.1) comprising of 61 compute nodes making up 1288 cores, 2176 GB RAM and 183 TB diskspace. It uses Open Grid Engine as a batch queuing system which schedules compute jobs and the corresponding requested resources. Compute jobs of Java, C/C++, Python, R and Matlab can be executed on our cluster. The cluster is also equiped with Hadoop File System (HDFS) and can be used for experiments involving distributed processing of large data sets across compute nodes. ISMLL members can access the Cluster Wiki

ISMLL's GPU Machine - Xeon-Phi

For massively parallel computations, our new server with 4 XeonPhi coprocessors has 128 GB RAM, 40 cores, 11.5 TB diskspace and 4 Xeon-Phi coprosessors with 60 cores and 4 threads per core, every coprocessor has 1011 gflops.