CyEnce consists primarily of 248 SuperMicro servers each with 16 cores, 128 GB of memory, GigE and QDR (40Gbit) Infiniband interconnects. Besides these compute nodes there are also 48 accelerator nodes, which are similar, but in addition contain either 2 NVIDIA Tesla K20 GPUs (in each of 24 nodes) or two 60 core Intel Xeon Phi Accelerator cards (in each of other 24 nodes). One large memory node contains 32 cores and 1 TB of main memory.
|Number of Nodes||Processors per Node||Cores per Node||Memory per Node||Interconnect||Local $TMPDIR Disk||Accelerator Card|
|248||Two 2.0 GHz 8-Core Intel E5 2650||16||128 GB||40G IB||2.5 TB||N/A|
|24||Two 2.0 GHz 8-Core Intel E5 2650||16||128 GB||40G IB||2.5 TB||Two NVIDIA K20|
|24||Two 2.0 GHz 8-Core Intel E5 2650||16||128 GB||40G IB||2.5 TB||Two 60 core Intel Phi 5110P|
|1||Four 2.0 GHz 8-Core Intel E5 2650||32||1 TB||40G IB||2.5 TB||N/A|
CyEnce has a 288TB shared Lustre scratch diskspace named /lustre, and 768TB of RAID-6 long term NFS disk space under quota. To ensure data integrity of the large NFS space, nightly rsyncs (backups) are performed to a read-only space within CyEnce, so that data can be recovered from the day before in case of filesystem corruption.