DCC: Retired December 2013
The DCC system was commissioned in its original form in 2010, and evolved to handle specialised processing workloads for climate science (for CMIP data from the Earth System Grid), genomics (for the analysis of data from sequencing machines), and earth observation (specifically the processing of satellite imagery).
DCC provided a mix of physical and virtual resources in the same cluster, with all nodes having direct access to the:
- /projects (via NFS), massdata, and the global Lustre (/g/data) filesystems.
Detailed System Configuration
DCC was a 192 core system comprising a single Dell M1000E blade chassis with:
- 16 Dell M710HD blades, each containing 2 sockets of Intel Xeon Westmere (6 cores each) and 144 GB of memory;
- Fast local disk;
- Fibre channel;
- 10 GigE networking.