Today Supermicro introduced new SuperServer solutions optimized for NVIDIA Tesla P100 accelerators with the new Pascal GPU architecture.
Our high-performance computing solutions enable deep learning, engineering, and scientific fields to scale out their 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,” said Charles Liang, President and CEO of Supermicro. “With our latest innovations incorporating the new NVIDIA P100 processors in a performance and density optimized 1U and 4U architectures with NVLink, our customers can accelerate their applications and innovations to address the most complex real world problems.”
Supermicro’s SuperServer 1028GQ-TXR(T) is a 1U server optimized to support four of the new NVIDIA Tesla P100 SXM2 accelerators for modern high bandwidth, low latency HPC clusters incorporating the latest NVIDIA NVLink GPU interconnect technology with over 5x the bandwidth of PCI-E 3.0. This SuperServer provides the best high performance parallel computing solution in a 1U form-factor for up to 4 Tesla P100 SXM2 accelerators with no GPU preheating design, which ensures uncompromised performance and stability under the most demanding workloads with up to 85 TFLOPS FP16 of peak performance.
Supermicro also debuts the density optimized 4U SuperServer 4028GR-TR(T)2 system supporting up to 10 PCI-E Tesla P100 accelerators for up to 187 TFLOPS FP16 peak performance with GPU Direct RDMA support. Supermicro’s innovative and GPU optimized single root complex PCI-E design is proven to dramatically improve GPU peer-to-peer communication efficiency over QPI, with up to 21% higher QPI throughput and 60% lower latency compared to previous generation products. Both 1U and 4U SuperServers support dual Intel Xeon processor E5-2600 v4/v3 product families, up to 3TB DDR4-2400MHz memory, optional dual onboard 10GBase-T ports, and redundant Titanium level (96%) digital power supplies.
Supermicro’s new high-density servers are optimized to fully leverage the new NVIDIA Tesla P100 accelerators to provide enterprise and HPC customers with an entirely new level of computing horsepower,” said Ian Buck, General Manager of the Accelerated Computing Group at NVIDIA. “The new SuperServers deliver superior energy-efficient performance for compute-intensive data analytics, deep learning and scientific applications while minimizing power consumption.”
With the convergence of Big Data Analytics, the latest GPU architectures, and improved Machine Learning algorithms, Deep Learning applications require processing power of multiple GPUs that must communicate efficiently and effectively to expand the GPU network. Supermicro’s single-root GPU system allows multiple GPUs to communicate efficiently to minimize latency and maximize throughput as measured by the NCCL P2PBandwidthTest.
In addition to the 4U supporting up to 10 Tesla P100s, Supermicro also offers the following SuperServers optimized for these new Pascal GPUs:
- 1U 4 Pascal GPU Optimized SuperServer solution – SuperServer 1028GQ-TR generates massively parallel processing power for Machine Learning applications.
- 2U 6 Pascal GPU Optimized SuperServer solution – SuperServer 2028GR-TRHT delivers the highest density massively parallel processing power for Machine Learning applications in 2U.
- 1U Ultra SuperServers – Designed to deliver unrivaled performance, flexibility, scalability, and serviceability, Supermicro’s 1028U servers support the new Pascal GPU. Ideal for demanding Enterprise workloads, these server support Intel Xeon processor E5-2600 v4 and v3 product families (160W/up to 22 Cores), up to 3TB of memory in 24 DIMMs, SATA3 with optional SAS3 and NVMe support for increased storage bandwidth, Ultra Riser options available which includes built-in 1G, 10GBase-T, 10G SFP+, 40G, and InfiniBand options, and Redundant Titanium Level (96%+) power supplies.
- 2U Ultra SuperServers – Designed to deliver unrivaled performance, flexibility, scalability, and serviceability, Supermicro’s 2028U SuperServers support up to 4 Pascal GPUs. Ideal for demanding Enterprise workloads, these server support Intel Xeon processor E5-2600 v4 and v3 product families (160W/up to 22 Cores), up to 3TB of memory in 24 DIMMs, SATA3 with optional SAS3 and NVMe support for increased storage bandwidth, Ultra Riser options available which includes built-in 1G, 10GBase-T, 10G SFP+, 40G, and InfiniBand options, and Redundant Titanium Level (96%+) power supplies.