FREMONT, CA — (Marketwired) — 11/18/13 — AMAX, a leading innovator of High Performance Computing (HPC), dynamic Enterprise IT, and custom solutions, today announced during the annual Supercomputing (SC13) show in Denver, CO, that it will offer the new NVIDIA® Tesla® K40 family of GPU accelerators in its and platforms. The 235W K40 accelerators outperform CPUs by up to 10x and include a Tesla GPUBoost feature that enables power headroom to be converted into user-controlled performance boost, as well as 12GB of GPU accelerator memory to process 2x larger datasets, making it ideal for big data analytics and large-scale scientific computations.
The ClusterMax and ServMax platform solutions featuring the new K40 accelerators are ideal for universities, national research labs, government, and enterprise high-performance computing environments where massive computational power is required. To meet these demands, AMAX offers up the ClusterMax SuperG, a complete turnkey GPU-enabled cluster that has been pre-configured, ready to deploy on-site and configurable with up to 108 NVIDIA Tesla K40 GPU accelerators in a standard 42U cabinet, delivering over 460+ Teraflops of peak single precision performance and 150+ Teraflops of peak double precision performance across 311,040 CUDA cores per standard 42U cabinet.
“With NVIDIA-s powerful and efficient Tesla K40 accelerators and AMAX-s dynamic cluster and server solution designs, we are able to deliver breakthrough performance and compute efficiency for double-precision focused applications,” said Julia Shih, Director of Strategic Alliance at AMAX. “With SMX (streaming multiprocessor) design, Dynamic Parallelism, Hyper-Q, GPUDirect with RDMA technologies embedded in the Kepler compute architecture, the performance increase we-ve seen in data analytics, satellite imaging, and weather modeling applications have simply been astounding.”
The latest AMAX solutions designed for the Tesla K40 accelerators include:
– :
Up to 311,040 GPU cores, 460+ Teraflops of single precision and 150+ Teraflops of double precision performance and 1,296 dedicated GPU memory per 42U cluster
Features FDR (56Gb/s) Infiniband interconnect and NVIDIA GPUDirect technology for high-speed cluster communication
Supports the latest Intel® Xeon® E5-2600 v2 series processors
Integrated Cluster Management to manage and monitor cluster & GPU resources
Off-the-shelf and ready-to-deploy K40 cluster solution with both hardware and software, delivering 55+, 110+, 220+ Teraflops of single precision, and 51+, 102+, 205+ Teraflops of double precision performance per cluster
Pre-configured with 4x, 8x, or 16x K40 GPU cluster nodes, with 12x, 24x, 48x GPUs per cluster
Dual Intel® Xeon® E5-2600 v2 series processors / up to 512GB DDR3 1866Mhz memory
Integrated Cluster Management to manage and monitor cluster & GPU resources
– :
2U rack server with support for 4x Tesla K40 accelerators ( >5+ Teraflops DP performance, 17+ Teraflops SP performance, 11,520 cores, 20GB of dedicated GPU memory )
Dual Intel® Xeon® E5-2600 v2 series processors
High-capacity storage with up to 8 hot-swap 3.5 SAS/SATA drive bays
Options for QDR/FDR Infiniband
“The Tesla K40 accelerator delivers tenfold application acceleration when paired with leading CPUs, and three times higher energy efficiency than previous-generation GPU accelerators,” said Sumit Gupta, general manager of Tesla accelerated computing at NVIDIA. “The new accelerators enable premier system integrators like AMAX to provide cluster and server solutions with previously unattainable levels of computing power.”
AMAX is showcasing a live demo of a Tesla K40 accelerator-enabled system for Supercomputing 2013 (booth # 4115), held at the Colorado Convention Center in Denver, CO. AMAX-s GPU solutions optimized for the new NVIDIA Tesla K40 GPU accelerators, including cluster and server solutions, are available immediately. A FREE GPU Test Drive cluster will be available for qualified applicants to remotely access an AMAX engineered, in-house Tesla K40 accelerator cluster. For more information on AMAX products and services, please visit: .
Founded in 1979, AMAX is a trusted leader in Datacenter, HPC and OEM Solutions in North America. Headquartered in Fremont, California, with locations in China to service the APAC region, AMAX is an open-architecture manufacturer specializing in innovative and scalable cluster, server, and storage products developed for Datacenter, HPC, Cloud and Big Data applications. AMAX-s mission is to provide best-in-class engineering, architectural design, manufacturing, global logistics & support to empower customers to solve the most complex computing challenges, handle massive amounts of data and meet product development demands to maintain a competitive advantage. As an Official OCP Solutions Provider, AMAX can build datacenter solutions using both standard and OCP architecture.
Image Available:
NAME: Julia Shih
Email Address:
Phone: 510-497-8828
You must be logged in to post a comment Login