Super Micro Computer, Inc., a global leader in high-performance, high-efficiency server, storage technology and green computing, exhibits its latest high-performance computing (HPC) solutions at the Supercomputing 2013 (SC13) conference this week in Denver, Colorado. In sync with the launch of the NVIDIA Tesla K40 GPU accelerator, Supermicro debuts new 4U 8x GPU SuperServer that supports the new and existing active or passive GPUs (up to 300 W) with an advanced cooling architecture that splits the CPU (up to 150 W x2) and GPU (up to 300 W x8) cooling zones on separate levels for maximum performance and reliability.

In addition, Supermicro has 1U, 2U, 3U SuperServers, FatTwin, SuperWorkstations and SuperBlade platforms ready to support the new K40 GPU accelerator. These high performance, high density servers support up to twenty GPU accelerators per system and in scaled out Super Clusters provide massive parallel processing power to accelerate the most demanding compute intensive applications. Supermicro's new platforms extend the industry's most comprehensive line of servers, storage, networking and server management solutions optimized for Engineering and Scientific Research, Modeling, Simulation and HPC supercomputing applications."Supermicro's HPC servers and solutions deliver the performance, scalability and reliability needed to answer the most complex challenges of our time," said Charles Liang, President and CEO of Supermicro. "Our extensive supercomputing solutions range from 5x GPU workstations to 6x GPUs in 1U/2U, 8x GPUs in 4U and 30x GPUs in 7U blade severs. Our GPU platforms are unrivaled in the industry and provide exactly the best configurations optimized for any scientific, engineering or big data analytics application. With the addition of a new 4U single node, 8x GPU server and a new 2U TwinPro platform, the HPC community can build even higher density compute clusters to deliver maximum parallel computing performance per watt, per dollar, per square foot."

"The Tesla K40 GPU accelerator provides double the memory and 10 times higher performance than today's fastest CPUs, enabling enterprise data center and HPC customers to solve their most complex engineering and big data analytics computing challenges," said Sumit Gupta, general manager of Tesla Accelerated Computing Products at NVIDIA. "When combined with Supermicro's high-density, scalable systems, the new Kepler-based accelerators deliver high performance computational horsepower with maximum energy efficiency."

Supermicro's new GPU accelerator-optimized server solutions on exhibit this week at SC13 include:

• NEW 4U 8x GPU SuperServer (SYS-4027GR-TR) - Ultra-high GPU density with massive parallel processing power in 4U form factor. System supports 8x NVIDIA Tesla K40, K20, K20X or K10 active or passive GPU accelerators (up to 300W) + additional full-height, full-length 2x PCI-E 3.0 x8 and 1x PCI-E 2.0 x4 slots, dual Intel Xeon E5-2600 v2 "Ivy Bridge" processors (up to 150W), 24x Reg. ECC DDR3 1600 MHz DIMM support (up to 768 GB), 2x 10GbaseT or GbE ports with 1x dedicated IPMI 2.0 port and 24x 2.5" hot-swap SAS/SATA/SSD bays. The 30" depth chassis features redundant Platinum Level high-efficiency 1600W power supplies (up to 4) and an advanced thermal cooling architecture with two rows of mid-chassis fans and separate CPU/GPU cooling zones.
• NEW 2U TwinPro (SYS-2027PR-DTR) / TwinPro² (SYS-2027PR-HTR) - Supermicro takes its 2U Twin architecture to the next level of performance, flexibility and expandability with the high efficiency 2-node TwinPro and high density 4-node TwinPro². Each node supports dual Intel Xeon E5-2600 v2 processors and the 2-node 2U TwinPro accommodates a NVIDIA Tesla GPU accelerator with support for two additional add on cards per node. The systems feature greater memory capacity up to 16x DIMMs, 12Gb/s SAS 3.0 support, NVMe optimized PCI-E SSD interface, additional PCI-E expansion slots, 10GbE and onboard QDR/FDR InfiniBand for maximized I/O.
• 1U SuperServer (SYS-1027GR-TRT2) - supports 3x GPUs, dual Intel Xeon E5-2600 series processors (up to 130W TDP), up to 512 GB memory in 16x DIMM slots and 4x hot-swap 2.5" SATA3 HDD bays. Features 1600W redundant Platinum Level high-efficiency (94%+) power supplies and smart server management tools.
• 1U SuperServer (SYS-1027GR-TQFT) - supports 4x GPUs, dual Intel Xeon E5-2600 series processors (up to 115W TDP), up to 256 GB memory and 4x hot-swap 2.5" SATA3 HDD bays. Features 1800W Platinum Level high-efficiency (94%+) power supplies and smart server management tools.
• 2U SuperServer (SYS-2027GR-TRFH) - supports 6x GPUs, dual Intel Xeon E5-2600 series processors (up to 130W TDP), up to 256 GB memory and 10x hot-swap 2.5" SATA HDD bays. Features redundant 1800W Platinum Level high-efficiency (94%+) power supply and smart server management tools.
• 3U SuperServer (SYS-6037R-72RFT+) - supports 2x GPUs (passive cooling with optional GPU fan kit installed), dual Intel Xeon E5-2600 v2 series processors (up to 135W TDP), up to 1.5 TB memory and 8x hot-swap 3.5" SAS2 HDD bays. Features redundant 1280W Platinum Level high-efficiency (94%) digital switching power supplies.
• 4U FatTwin (SYS-F627G3-FT+ / G2-FT+) - 4x hot-plug nodes supporting 12x GPUs (3x per node) and dual Intel Xeon E5-2600 series processors (up to 130W TDP) per node. Available with front I/O and 2x 3.5" or 6x 2.5" hot-swap HDD bays. Features redundant 1620W Platinum Level high-efficiency (94%+) power supplies.
• 4U/Tower SuperWorkstation (SYS-7047GR-TRF / -TPRF) - Ultimate performance (NVIDIA Maximus Technology Certified) and expandability with support for up to 5x GPUs, dual Intel Xeon E5-2600 series processors, up to 512 GB memory and 8x hot-swap 3.5" HDD bays. Tower/4U rackable chassis features redundant 1620W Platinum Level high-efficiency (94%) power supplies.
• SuperBlade Solutions - The all-in-one 7U SuperBlade features redundant Platinum Level high-efficiency (94%+) power supplies, high speed connectivity through network switch modules, including 56Gb/s FDR IB (SBM-IBS-F3616M), FC/FCoE (SBM-XEM-F8X4SM), 10GbE (SBM-XEM-X10SM) and 1/10GbE (SBM-GEM-X3S+) and centralized remote management software.
  3x GPU SuperBlade (SBI-7127RG3) - Supports 3x NVIDIA Tesla K20X GPUs in the SXM form factor, dual Intel Xeon E5-2600 series processors, up to 256 GB memory and onboard BMC for IPMI 2.0 support. 10x blades in 7U SuperBlade enclosure scales to best density (180x GPUs and 120x CPUs) and performance (256 TFLOPS theoretical) per 42U rack.
  2x GPU SuperBlade (SBI-7127RG-E) - Supports 2x GPUs, dual Intel Xeon E5-2600 series processors, up to 256 GB memory, 1x SSD or 1x SATA-DOM, and onboard BMC for IPMI 2.0 support. 10x blades in 7U SuperBlade enclosure offers high density (120x GPUs and 120x CPUs) and performance (178 TFLOPS theoretical) per 42U rack.

Experience Supermicro's latest High Performance Computing solutions this week at Supercomputing 2013 in Denver, Colorado in booth #3132 at the Colorado Convention Center, November 18th - 22nd. Supermicro server solutions optimized for NVIDIA Tesla K40 GPU accelerators are also on exhibit at NVIDIA's SC13 booth #613. For information on Supermicro's complete line of GPU enabled supercomputing platforms visit www.supermicro.nl/GPU.