Rossmann

Overview of Rossmann

Rossmann is a compute cluster operated by ITaP, and is a member of Purdue's Community Cluster Program. Rossmann went into production on September 1, 2010. It consists of HP (Hewlett Packard) ProLiant DL165 G7 nodes with 64-bit, dual 12-core AMD Opteron 6172 processors (24 cores per node), either 48 GB or 96 GB of memory, and 250 GB of local disk for system software and local scratch storage. Nodes with 192 GB of memory and either 1 TB or 2 TB of local scratch disk are also available. All nodes have 10 Gigabit Ethernet interconnects.

Namesake

Rossmann is named in honor of Michael Rossmann, Purdue's Hanley Distinguished Professor of Biological Sciences. More information about his life and impact on Purdue is available in an ITaP Biography of Michael Rossmann.

Detailed Hardware Specification

Rossmann consists of five logical sub-clusters, each with a different memory/storage configuration. All nodes in the cluster have dual 12-core AMD Opteron 6172 processors and 10 Gigabit Ethernet (10GigE).

Sub-Cluster Number of Nodes Processors per Node Cores per Node Memory per Node Interconnect Disk
Rossmann-A 393 Two 2.1 GHz 12-Core AMD 6172 24 48 GB 10 GigE 250 GB
Rossmann-B 40 Two 2.1 GHz 12-Core AMD 6172 24 96 GB 10 GigE 250 GB
Rossmann-C 2 Two 2.1 GHz 12-Core AMD 6172 24 192 GB 10 GigE 1 TB
Rossmann-D 4 Two 2.1 GHz 12-Core AMD 6172 24 192 GB 10 GigE 2 TB
Rossmann-H 11 Two 2.1 GHz 12-Core AMD 6172 24 48 GB 10 GigE 8 TB

Rossmann nodes run Red Hat Enterprise Linux 5 (RHEL5) and use Moab Workload Manager 6 and TORQUE Resource Manager 3 as the portable batch system (PBS) for resource and job management. Rossmann also runs jobs for BoilerGrid whenever processor cores in it would otherwise be idle. The application of operating system patches occurs as security needs dictate. All nodes allow for unlimited stack usage, as well as unlimited core dump size (though disk space and server quotas may still be a limiting factor).

For more information about the TORQUE Resource Manager:

On Rossmann, ITaP recommends the following set of compiler, math library, and message-passing library for parallel code:

  • PGI 11.8
  • ACML
  • OpenMPI 1.4.4

To load the recommended set: 

$ module load devel

To verify what you loaded: 

$ module list
About Michael Rossmann