TYAN Displays HPC and Cloud Computing Server Platforms Featuring Upcoming Intel Xeon Processor Scalable Family at Computex 2017
TAIPEI, Taiwan, May 30, 2017 /PRNewswire/ -- TYAN®, an industry-leading server platform design manufacturer and subsidiary of MiTAC Computing Technology Corporation, exhibits its new line-up of HPC, cloud computing and storage server platforms this week at Computex 2017 in Taipei, Taiwan. TYAN showcases a full lineup of platforms based on the upcoming Intel® Xeon® Processor Scalable Family that are targeted at the Data Center, Virtualization, Supercomputing, Enterprise and Embedded infrastructure markets.
"With increased needs of data-intensive applications driven by the growth of artificial intelligence, virtual reality and high-performance computing, customers need an advanced solution to address the demands of big-data, in-memory workloads for their server infrastructures," said Danny Hsu, Vice President of MiTAC Computing Technology Corporation's TYAN Business Unit. "TYAN plans to offer latest server platforms based on the Intel Xeon Processor Scalable family from mid-2017, and promises to offer workload-optimized performance, power efficiency, and platform features."
High-density GPU and Intel® Xeon Phi™ Coprocessor Platforms Optimized for HPC and Machine Learning Applications
TYAN's next generation HPC computing platforms are all based on the Intel Xeon Scalable Processor Family and are designed for the heavy computing workloads of big data and high performance data analysis applications. To meet these demands, TYAN have three platforms on display that target the HPC, Machine Learning, and Technical Computing markets.
The FT77D-B7109 is a 4U dual root complex GPU server with two CPU sockets and support for up to 8 Intel Xeon Phi coprocessor x200 series cards. The platform specializes in embarrassingly parallel workloads including scientific computing, genetic sequencing, oil & gas discovery, large scale facial recognition, and brute force cryptography.
TYAN is also displaying a new high performance workstation platform named the FT48T-B7105. This workstation gives maximum I/O to the professional power user, with support for up to 5 Intel Xeon Phi coprocessors and is aimed at Digital Content Creation, Computer-Generated Imagery (CGI), and Computer-Aided Design (CAD) applications. TYAN's GA88-B5631 is a fully peer-to-peer single root complex 1U GPU server. Featuring a single Intel Xeon Scalable Processor Family CPU socket, the platform supports up to 4 Intel Xeon Phi coprocessor, and is ideal for many of today's emerging cognitive computing workloads such as Machine Learning and Artificial Intelligence.
TYAN Highlights Extreme Performance, Density and Scalability for Next-Generation Datacenter, Enterprise and Cloud
Powered by the latest Intel Xeon Processor Scalable Family, TYAN's new range of cloud computing and storage platforms are optimized for data intensive workloads and virtualization applications to deliver extreme performance, density and scalability with power and cost efficiency.
The 1U GT75B-B7102 with support for 10 2.5" small form factor SATA bays, four of which can support NVMe U.2 drives, is an ideal platform for virtualization and in-memory databases. TYAN's 1U GT62F-B5630 is designed for hybrid NVMe/SATA cache storage with support for up to 8 hot-swap NVMe U.2 drives along with an OCP v2.0 LAN Mezzanine. The single CPU socket design makes it an ideal platform for workloads that work best within a single NUMA domain and require large amounts of high-speed flash, such as many media streaming applications.
The all-new TN200-B7108-X4S is a dual-socket 2U 4-Node all-flash server platform with support for 24x 2.5" NVMe U.2/SATA drives. Each node gets 8 NVMe drives across 6 PCIe x4 NVME U.2 hot-swap drive bays up front and a pair of internal 2280/22110 NVMe M.2 ports. With a total of 8 CPU sockets and 32 NVMe devices across the entire chassis, the TN200-B7108-X4S is an ideal platform for High Performance Computing workloads and hyper-converged all-flash storage applications.
Share this article