Announcements

2 Jan 2017

DAS-5/VU has been extended with 4 TitanX-Pascal GPUs. Check DAS-5 special nodes for an overview of all DAS-5 special nodes.

9 Nov 2016

CUDA 8 is now available on all DAS-5 sites with GPU nodes. Check the DAS-5 GPU page for usage info.

May, 2016

IEEE Computer publishes paper about 20 years of Distributed ASCI Supercomputer. See the DAS Achievements page.

28 Sep 2015

DAS-5/VU has been extended with 16 GTX TitanX GPUs.

6 Aug 2015

DAS-5/UvA has been extended with 4 GTX TitanX GPUs: two on both node205 and node206.

6 Jul 2015

DAS-5 is fully operational!

DAS-5 clusters


DAS-5 includes roughly 200 dual-eight-core compute nodes (primarily ASUS nodes with Intel E5-2630v3 CPUs), spread out over six clusters, located at five sites. The system has been built by ClusterVision. Like its predecessor, DAS-4, DAS-5 has a mostly homogeneous core:

Cluster Nodes Type Speed Memory Storage Node HDDs Network Accelerators
VU 68 dual 8-core 2.4 GHz 64 GB 128 TB 2*4TB IB and GbE 16*TitanX+ Titan/GTX980/K20/K40
LU 24 dual 8-core 2.4 GHz 64 GB 128 TB 2*4TB IB and GbE -
UvA 18 dual 8-core 2.4 GHz 64 GB 128 TB 2*4TB IB and GbE 4*TitanX
TUD 48 dual 8-core 2.4 GHz 64 GB 192 TB 2*4TB IB and GbE -
UvA-MN 31 dual 8-core 2.4 GHz 64 GB 128 TB 2*4TB IB and GbE
ASTRON 9 dual 8/10/14-core 2.6 GHz 128/512 GB 96 TB 2*4TB IB,
40 GbE, and GbE
NVIDIA/AMD GPUs; Xeon Phi

However, every DAS-5 site also some special hardware extensions that focus on local research priorities.

Besides using the ubiquitous Ethernet (1 Gbit/s at the compute nodes, 10 Gbit/s on the head nodes), DAS-5 also employs the high speed InfiniBand interconnect technology. FDR InfiniBand (IB) is used as an internal high-speed interconnect. To connect with the other DAS-5 clusters, a second ethernet adapter of every compute node is used that can communicate over dedicated Ethernet lightpaths by means of an OpenFlow switch at every site. These lightpaths are implemented via a fully optical (DWDM) backbone in the Netherlands, SURFnet7. See also: DAS-5 Connectivity.

The operating system the DAS-5 runs is CentOS Linux. Cluster management is done using Bright cluster management suite. In addition, software from many sources is available to support research on DAS-4: the SLURM resource management system, various MPI implementations (e.g., OpenMPI and MPICH), optimizing compilers, visualization packages, performance analysis tools, etc.

DAS-5 is a very actively used research tool for experimental Computer Science in the Netherlands, as also witnessed by the special hardware (new node types, accelerators, storage) that has been added to the above list since the start of the project.