New Servers Based on AMD EPYC and AMD Radeon Instinct Platforms Appear at SC17
At SuperComputing 17 this week, the enterprise computing conference, AMD is announced that it has ramped production of the EPYC processors and it has several OEMs ready, distributors ready, and system integrators expanding their portfolios.
AMD combines this portfolio with software, featuring the new ROCm 1.7 open platform with updated development tools and libraries, enabling complete AMD EPYC-based PetaFLOPS systems.
OEMs with EPYC enabled systems include ASUS, BOXX, GIGABYTE, HPE (Hewlett Packard Enterprise), Penguin Computing, Supermicro and Tyan. ASUS' RS720A-E9 and RS700A-E9 1U/2U servers are tergetting HPC and Virtualization, the BOXX system combines EPYC with Radeon Instinct for multi-GPU compute solutions and deep learning, HPE tergets complex workloads and Supermicro is moving from tower form factors to 1U, 2U and 4U for HPC and storage.
By supporting both heterogeneous supercomputing systems and memory-bound, CPU driven platforms with EPYC, AMD addresses the needs of multiple workloads. The company claims that the EPYC 7601 systems offer up to a 3X advantage in performance per dollar vs the Intel Xeon Platinum 8180M.
A highlight for AMD at SC17 is the Inventec P47 system that combines the performance of a single EPYC 7000 series CPU with four Radeon Instinct MI25 GPUs, each delivering up to 12.3 TFLOPS of single precision performance in a scalable platform. For large to hyperscale deployments, AMAX has developed the [SMART]Rack P47, an all-inclusive high-performance rackscale appliance featuring 20x P47 platforms to provide up to a PetaFLOPS of single precision compute performance and more than 10 terabytes of DDR4 memory per rack. The [SMART]Rack P47 also features HPC-optimized [SMART]DC DCiM software to remotely monitor, manage and orchestrate GPU-based deployments where real-time temperature, power and system health are particularly crucial to help ensure uninterrupted operation, as well as AMD's ROCm software platform for the easy implementation for deep learning, inference and training workloads.
AMD EPYC and AMD Radeon Instinct performance is fully supported by the new ROCm 1.7 release. Expanding on the open source software platform for heterogeneous computing systems, ROCm 1.7 delivers math libraries and software development support using modern programming languages to unlock the power of GPU acceleration and other accelerators like FPGAs. The ROCm 1.7 release includes multi-GPU support for the latest Radeon GPU hardware, as well as support for TensorFlow and Caffe in the MIOpen libraries.