TOKYO, JAPAN: Supermicro announces the release of various SuperServer solutions equipped with Pascal GPU architecture. The new server leverages NVIDIA Tesla P100 to provide high-performance in compute-intensive data analytics, deep learning, and scientific applications.
Supermicro’s SuperServer 1028GQ-TXR (T) supports four of the new NVIDIA Tesla P100 SXM2 accelerators including the NVIDIA NVLink GPU interconnect technology with over 5x the bandwidth of PCI-E 3.0. The 1U server also provides low latency HPC clusters and parallel computing solution in 1U form-factor for up to 4 Tesla P100 SXM2 accelerators.
The 4U SuperServer 4028GR-TR (T)2 system supports up to 10 PCI-E Tesla P100 accelerators enabling higher performance with GPU Direct RDMA support. The latest design comes with 21 percent higher QPI throughput and 60 percent lower latency compared to the previous models.
Supermicro also offers SuperServers for other Pascal GPUs such as SuperServer 1028GQ-TR, SuperServer 2028GR-TRHT, 1U Ultra SuperServers, and 2U Ultra SuperServers.
The convergence of Big Data analytics has effectively expanded the GPU network for improved Machine Learning and Deep Learning applications. The single-root GPU system reduces the latency to enable multiple GPUs communicate efficiently.
“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,” says 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.”