Custom configurations

Many custom configurations are available, including drive configurations, expansion cards, etc.

If you don't see the configuration you're looking for, use your Custom System form so we can configure your perfect machine.

Customer testimonial

"By the way, Jason has been a great help. The testing would not have happened without him. Not only did he install MPI for me but emacs as well. He deserves a raise. Thanks again for all you have done. " -- Buff M. at SAIC

Find out even more reasons to purchase from us

 

Warranty and support

All systems come with our standard 9x5 one year cross-ship return to depot warranty.  Upgraded warranty options including multiple years, 24x7, and on-site service are available.  Read more information about our warranties.

Lifetime technical support is included with every Advanced Clustering system.  This includes support via, email, or our toll-free support telephone line.

Home Products GPU systems GPU clusters

GPU Clusters

2u-blade-angle-250pxAdvanced Clustering's GPU clusters consist of our innovative Pinnacle Flex Compute Blade products and the nVidia Kepler K20 line of GPUs. This flexible and modular design allows for mixing and matching of GPU and CPU configurations while at the same time preserving precious rack and datacenter space.

The most common configuration is a pair of Pinnacle Flex Compute blade servers, each with one K20 GPU in only 2U of space.  Each server system can be customized with either AMD or Intel processors, multiple RAM configurations, and gigabit ethernet with options for 10/40-gigabit ethernet or InfiniBand.

 

Features and Benefits of Kepler K20 GPU

Tesla K20 GPUs Powered by the Kepler-Generation of the CUDA Architecture
Delivers cluster performance at 1/10th the cost and 1/20th the power of CPU-only systems based on the latest multi-core CPUs.

2,496 CUDA Cores
Delivers up to 1.17 Teraflops of double-precision peak performance in each GPU, enabling 2 Teraflops of double-precision performance in 2U of space. Single precision peak performance is more than 3.5 Teraflops per GPU.

ECC Memory
Meets a critical requirement for mission-critical applications with uncompromised computing accuracy and reliability. Offers protection of data in memory to enhance data integrity and reliability for applications. Register files, L1/L2 caches, shared memory, and DRAM all are Ecc protected.

5GB of GDDR5 Memory per GPU
Maximizes performance and reduces data transfers by keeping larger data sets in local memory that is attached directly to the GPU. 

CUDA Programming Environment
Choose OpenACC, CUDA toolkits for C, C++, or Fortran to express application parallelism and take advantage of the innovative Kepler architecture.

GPU Computing Applications
Seismic processing, CFD, CAE, Financial computing, Computational Chemistry and Physics, Data analytics, Satellite imaging, Weather modeling.

 

Features and Benefits of Kepler K40 GPU

Tesla K20 GPUs Powered by the Kepler-Generation of the CUDA Architecture
Delivers cluster performance at 1/10th the cost and 1/20th the power of CPU-only systems based on the latest multi-core CPUs.

2,880 CUDA Cores
Delivers up to 1.17 Teraflops of double-precision peak performance in each GPU, enabling 2 Teraflops of double-precision performance in 2U of space. Single precision peak performance is more than 3.5 Teraflops per GPU.

ECC Memory
Meets a critical requirement for mission-critical applications with uncompromised computing accuracy and reliability. Offers protection of data in memory to enhance data integrity and reliability for applications. Register files, L1/L2 caches, shared memory, and DRAM all are Ecc protected.

12GB of GDDR5 Memory per GPU
Maximizes performance and reduces data transfers by keeping larger data sets in local memory that is attached directly to the GPU. 

CUDA Programming Environment
Choose OpenACC, CUDA toolkits for C, C++, or Fortran to express application parallelism and take advantage of the innovative Kepler architecture.

GPU Computing Applications
Seismic processing, CFD, CAE, Financial computing, Computational Chemistry and Physics, Data analytics, Satellite imaging, Weather modeling.

Features Tesla K40 Tesla K20
Number and type of GPU 1x Kepler GK110B 1x Kepler GK110
Peak DP floating point performance 1.43 Tflops 1.17 Tflops
Memory bandwidth (ECC on) 269 GB/sec 194 GB/sec
Memory size (GDDR5) 12GB 5GB
CUDA cores 2880 2496

More information

Contact Us

Whatever your GPU computing needs are, you can be sure that Advanced Clustering has the experience and technical know-how to help. Our expert staff can assist you with your GPU computing questions and design a solution that's right for you. Give us a call at 866-802-8222 or This e-mail address is being protected from spambots. You need JavaScript enabled to view it for more information.