Supermicro Releases Broad Range of HPC SuperServer, Storage, and Networking Solutions to Exploit Deep Learning
FREMONT, CA: Supermicro, a server and Storage solutions provider, announces the release of a wide range of High Performance Computing (HPC) solutions, engineered by it.
The new range of Supermicro’s HPC solutions enable users in fields such as deep learning, engineering, and scientific studies, take full advantage and scale out compute clusters to accelerate their most demanding workloads and achieve fastest time-to-results with maximum performance.
The company launches latest innovations in HPC architectures and technologies including:
SuperServer supporting Intel Xeon Phi processor
2U 4-Node server supporting new Intel Xeon Phi processors with integrated or external Intel Omni-Path fabric option, together with associated 4U/Tower development workstation.
2U 40 Dual-port NVMe All-Flash SuperStorage Solutions
A 2U 40 dual-port, dual-controller, all- NVMe system supporting up to 30GB/s per system throughput via Intel 100G Omni-Path networking, unrivaled in the industry.
2U Simply Double SuperStorage
Solutions offer up to twice the storage capacity and IOPS in the same amount of space of traditional 2U front load storage systems. Second set of drive bays are arrayed in a patented Riser Bay located on the top of the Simply Double systems for easy access and servicing. Available in 2.5" or 3.5" drive bays, and that also support All-Flash NVMe SSDs or SAS 3.0 HDDs.
SuperServer supporting GPU
Accelerator with next generation architecture-based Solutions – Supermicro’s Enterprise-class High Performance SuperComputing solutions are available in the industry’s broadest form factors and highest density (1U, 4 next generation P100 GPU SYS-1028GQ-TXR/-TXRT) generating massively parallel processing power and unrivaled GPU peering via 80GB/s NVLINK for Machine Learning applications.
7U 8-Way (MP)SuperServer
Latest generation Supermicro 8-Way multi-processor (MP) system features up to 192 CPU cores and 192 DDR4 memory DIMM slots. Eight CPU modules and 2 storage modules are located in the front of the system with each CPU module supporting Intel Xeon processor E7-8800 v4/v3 product family with QPI up to 9.6GT/s, 24 DDR4 memory DIMMs.
Modules support latest Intel Xeon Processor E5-2600 v4 product family and are available with 20 Intel Xeon Phi coprocessor or GPU Blades; 2 coprocessor cards per blade server (SBI-7128RG-X/-F/-F2), 3 GPU per blade server (SBI-7127RG3), Data Center Blade (SBI-7428R-C3N, SBI-7428R-T3N), TwinBlade (SBI-7228R-T2F/-T2F2/-T2X), Storage Blade with NVMe support (SBI-7128R-C6N) solutions.
The MicroBlade enclosure can incorporate 1 Chassis Management Module, and up to 2x 10/2.5/1GbE SDN switches in 3U or up to 2 Chassis Management Modules, and up to 4 SDN Switches in 6U for efficient, high-bandwidth communications.
“With Supermicro HPC solutions deep learning, engineering, and scientific fields can scale out compute clusters to accelerate their most demanding workloads and achieve fastest time-to-results with maximum performance per watt, per square foot, and per dollar,” says Charles Liang, President and CEO of Supermicro. “With our latest innovations incorporating Intel Xeon Phi processors in a performance and density optimized Twin architecture, 8-socket scalable servers, 100Gbps OPA switch for high bandwidth connectivity, and high-performance NVMe for Lustre based storage, our customers can accelerate their applications and innovations to address the most complex real world problems.”