Institute Computational Science & Technology Resources

 

Computational Science

Computational science comprises software engineering related to modeling, simulation, visualization and data analytics applied to various applications domains such as bioinformatics, computational fluid dynamics, and finite element analysis.  In this category the Institute provides licensed analytics software resources and professional engineering services to assist users in their pursuit of solutions to their engineering, research and development problems.

Computational Technology

As part of its ongoing commitment to development, operation and maintenance of a regional cyberinfrastructure (often referred to as the "Milwaukee Metropolitan Grid"  MGrid), the Institute acquired and, beginning in April of 2009, has made available a publically accessible High Performance Computer (HPC) cluster, designated hpc01.mkei.org, and a large memory Visualization Server containing two GPU's.  These resources provide the platform for hosting high-fidelity modeling, simulation and visualization applications and services for regional .edu, .com and .gov institutions.
 

Dell HPC Cluster

Designated "hpc01.mkei.org"

 

hpc01 was originally launched in two phases, included 16 Dell PowerEdge 1950 servers, each with dual quad-core Intel Xeon E5440 proccessors, and 13 Dell PowerEdge r410 servers, each with dual quad-core Intel Xeon E5520 processors for a total of 29 nodes and 232 cores.  In 2013 it was upgraded with 16 Dell PowerEdge r620 servers, each with dual eight-core Intel Xeon E5-2660 [Sandybridge] processors, and 1 Dell PowerEdge r820 Deep Memory Node with 768 GB of RAM and four eight-socket Intel Xeon E5-4660 (Sandybridge-E) processors.  This brings hpc01 to 46 servers and 520 cores.

 

 

Visualization GPGPU/Cuda Server

Desgnated "mkvisual1.mkei.org"

mkvisual1 is a 2U server that contains dual six-core Intel Xeon X5650 [Westmere] processors, an NVIDIA Quadro 4000 (256-core) ideal for rendering high fidelity images, and an NVIDIA Tesla C2075 (448-core) for use with CUDA programming for a total of 12 CPU cores and 704 GPGPU cores.