More TFLOPS, Fewer Watts - Microway BioStack™-LS
Modular, Extendable, Managed Tesla GPU Cluster for Life Sciences
View other Microway Tesla Clusters
The Microway BioStack-LS is a modular, extensible cluster for life science researchers. It provides dense compute power that's easy to use. With over 49 TFLOPS of InfiniBand-connected computing power spread over 34,944 NVIDIA Tesla SMX GPU cores and 112 Intel Xeon CPU cores, the BioStack-LS provides performance vastly superior to traditional compute clusters. To suit life science customer needs, BioStack- LS is delivered fully tested, integrated, and ready to run NVIDIA's Bio WorkBench applications with the latest CUDA environment.
A core 9U configuration is available, yet BioStack also scales to multiple 42U cabinets for massive processing requirements.
NVIDIA Bio WorkBench Applications
- AMBER support with 10x speedup, including for MPI applications
- Molecular dynamics: NAMD, VMD, GROMACS, OpenMM, ACEMD, TeraChem, LAMMPS
- Bio-Informatics: GPU-HMMER, CUDA-BLASTP, MUMmerGPU, OpenEYE ROCS, CUDASW++
Bright Cluster Manager
Easy professional management with:
- Optimization for top throughput with continual health checks, automatic sidelining of troubled nodes
- Graphical displays of cluster utilization, open resources
- Low resource footprint: single daemon and database drives all processes
Microway's BioStack™ awarded
Best of Show at Bio-IT World Conference, April 2011
Implement Tesla Fermi GPUs for supercomputer performance in your workgroup. The BioStack-LS fully integrates:
- 16.38 TFLOPS double precision GPU performance (49.28 TFLOPS single)
- Up to 112 CPU Cores and 34,944 GPU cores in 9U
- Seven GPU Compute Nodes: each with 2 Tesla Kepler K20 2496 core GPUs and 16 CPU cores
- Intel Xeon CPU Cores at up to 2.9Ghz
- Mellanox 36 port ConnectX-3™ FDR InfiniBand switch
- Compact Cabinet for excellent scalability
- Bright Cluster Manager™ with CentOS, Red Hat, or SUSE Linux
- Choice of NVIDIA Bio Workbench applications installed and configured
NVIDIA Tesla based clusters like BioStack-LS deliver supercomputing performance at a lower power, lower cost, and using many fewer systems than standard CPU-only clusters. Successfully deployed in demanding applications at research institutes, universities, and enterprises, NVIDIA Tesla powers the most powerful supercomputer worldwide.
Please contact our Technical Sales Department to learn more about customizing Microway's GPU Compute Clusters to your needs.
Request a Quote