ASUS Announces NVIDIA-Certified ESC4000A, RS720A-E11, RS720-E10 Servers for AI Applications

Update : 2021/06/01 16:32:56

Taipei, Taiwan, 1 June, 2021 —ASUS, the leading IT company in servers and workstations that tops the industry in server performance with over 743 benchmark world records on SPEC.org , today announced its rack server and GPU server lineup is set to offer expanded support for both the NVIDIA® EGX® platform and NVIDIA BlueFieldⓇ-2 DPUs, with NVIDIA-Certified Systems™ either already available or pending with ESC4000A, RS720A-E11, and RS720-E10.

 

These NVIDIA-Certified Systems will enable the use of a broader range of applications for data center and enterprise segments, with harmonized hardware design. New ASUS server series that will be revealed later this year will extend the support for these NVIDIA-powered solutions, empowering customers to identify and implement the right system for diverse use cases and scenarios.

 

Specifically, the NVIDIA-Certified ASUS ESC4000A, RS720A-E11, RS720-E10 server models built on the NVIDIA EGX platform empower enterprise customers to speed up digital transformation in daily operations as well as AI workloads for AI training and AI-powered inferencing, plus virtualization and cloud applications. ASUS intends later in the year to expand its server lineup still further to embrace NVIDIA-Certified BlueField-2 systems, in addition to the existing and new NVIDIA-Certified servers.

 

ASUS NVIDIA-Certified servers

The latest ASUS ESC4000A-E10, RS720-E10 and RS720A-E11 servers are all NVIDIA-Certified, and feature both the NVIDIA A100 Tensor Core GPU and NVIDIA ConnectX®-6 HDR 200Gb/s InfiniBand adapters and Ethernet NICs. These systems conform to NVIDIA’s design best practices and have passed a set of certification tests that validate the best system configurations for performance, manageability, scalability and security. Part of the NVIDIA EGX platform , they enable enterprises to meet enterprise requirements for a variety of demanding workloads, including running NVIDIA AI Enterprise software for AI in virtualized environments, supporting remote collaboration and accelerated graphics in NVIDIA Omniverse™ Enterprise, as well as advancing data science.

 

ASUS NVIDIA-Certified BlueField-2 DPU-based servers

Later this year, ASUS will also debut all-new servers featuring NVIDIA BlueField-2 DPUs, or data-processing units, to enable breakthrough networking, storage and security performance. The systems will be submitted for NVIDIA certification as they come to market.

 

The inclusion of BlueField-2 enables a new set of capabilities to offload, accelerate and isolate networking, security and storage services, plus speed up data movement for AI, HPC, and streaming-video workloads. By offloading tasks from the CPU, a single BlueField-2 DPU can provide the same data-center infrastructure services that could require more than 100 CPU cores, freeing up server CPU cycles to focus on business-critical applications. Developers can rapidly create software-defined, hardware-accelerated, data center applications and services featuring BlueField-2 DPUs by using the NVIDIA Data Center Infrastructure-on-a-Chip Architecture (DOCA) SDK.

 

Complete lineup of ASUS NVIDIA-Certified servers is on the way

ASUS is committed to launching a complete lineup of NVIDIA-Certified Systems and BlueField-2 DPU-based servers, including both rack and GPU servers, during the second half (H2) 2021 and the first half (H1) of 2022. This pledge provides customers looking for close alignment with NVIDIA solutions the confidence needed to make their investment in ASUS servers and server equipment.

 

ASUS is also announcing support for NVIDIA Omniverse Enterprise, a simple to deploy, end-to-end 3D collaboration and true-to-reality simulation platform, optimized to run on NVIDIA RTX™ professional systems. Omniverse Enterprise transforms complex 3D workflows, enabling new heights of 3D production quality, unlimited iterations, and faster time-to-market.

 

ASUS has a long and proud track record of cultivating in AI and enterprise solutions that implement exclusive technologies, including ASUS Adaptable Topology and ASUS Performance Boost, to ensure the perfect balance between performance and power consumption. With its current and future NVIDIA-Certified EGX and BlueField-2 systems, ASUS will deliver more workload-oriented solutions for specific applications – such as high-performance computing (HPC), AI training and inferencing, and virtualization.