Hardware

The Computational Biology Core provides HPC resources for both the teaching and research mission.  The BBC Cluster is dedicated to the teaching mission and instructors should contact the core in advance of offering a course that will require students accounts and software.  

The new Xanadu cluster is accessible for all affiliates of UConn for research.   To request an account on the cluster, please contact us.  If you are planning a project with the CBC (or the CGI) and need to include a description of the computational resources, please use this text.

Teaching HPC - BBC Cluster (bbcsrv3.biotech.uconn.edu)

32-node Dell Linux cluster running Centos 6.3 and Rocks 6.1 x64.

Node configuration:

  • 1: 16 x Quad-core 2.5 Ghz AMD 6370P processors with 512 GB RAM
  • 4: 4 x 8-core 2.00 GHz Intel Xeon processors with 64 GB RAM
  • 17: 2 x Quad-core 2.53 GHz Intel Xeon processors with 32 GB RAM
  • 10: 2 x Dual-core 2.8 GHz AMD processors with 16 GB RAM

Local storage capacity: Dell PowerVault MD1000 18TB RAID Array

Research HPC - Xanadu Cluster (xanadu-submit-ext.cam.uchc.edu)

Node Configuration:

  • 2,400 CPU cores with 5.8 TB RAM
  • 7,000 GPU cores with large CPU-only and hybrid compute clusters + OSG cluster

Virtualization Infrastructure:

  • 456 CPU cores, 1 TB RAM VMWare server and desktop virtualization hosts hosting 100+ Windows/Linux virtual machines with SSD high OPS performance cache tier

Datacenter Infrastructure:

  • UPS generator backed power with redundant cooling
  • 3x40 Gbe dark finer connection to off-site DR location

Network (100+ Gbe):

  • Full non-oversubscribed 10/40 GbE datacenter network core layer
  • BioScienceCT Research Network – 100 GbE to CEN, Internet2, Storrs
  • New HPC Science DMZ – low latency, 80 Gb-capable firewall

Storage (2+ PB):

  • 394 TB EMC2 Isilon and 832 TB Qumulo shared scale-out clusters along with 1,056 TB Amplistor on-premise cloud storage

cern