Sharp Corporation will get in the AI server service by accepting orders for AI servers geared up with NVIDIA innovation in Japan beginning today, September 15, from corporations, city governments, research study organizations, information center operators, and other consumers. Ahead of the scheduled business launch in financial 2027, Sharp goals to resolve the growing need for AI facilities driven by the expansion of generative AI. The very first item offered for order is the AI Server for Enterprise Workloads (NVIDIA RTX Pro Server). Sharp likewise prepares to start accepting orders for the AI Server for Large-Scale Training and Inference (NVIDIA HGX System) at a later date.
In establishing its AI server organization, Sharp is developing out its sales structure along with its operation and upkeep structure following implementation. Sharp has actually respectively reached a fundamental contract on partnership with 2 sales representatives, Daiwabo Information System Co., Ltd.( head workplaces: Kita-ku, Osaka City and Minato-ku, Tokyo )and Ryoyo Ryosan, Inc. (head workplace: Chiyoda-ku, Tokyo). In addition, Sharp will supply upkeep assistance in partnership with TOMORROW NET Co., Ltd. (head workplace: Shinagawa-ku, Tokyo), which will be accountable for operation and upkeep.
Through cooperation with these partners, Sharp will construct a one-stop structure covering consumer propositions, implementation, operation, and upkeep, and will support the advancement and growth of AI facilities in Japan.
An AI server is a high-performance server that manages AI training and reasoning processing. As generative AI ends up being more extensive and information usage ends up being more advanced, there is growing need in Japan from corporations, city governments, research study organizations, and others who want to use AI on-premises (* 1) while guaranteeing security. At the exact same time, the worldwide supply-demand balance for essential elements such as GPUs (* 2) stays constrained, and obtaining AI servers can need substantial time and coordination.
In this environment, Sharp will react to market requires through partnership with Hon Hai Technology Group (Foxconn, head workplace: New Taipei City, Taiwan), which masters procurement and production.
Introduction of the AI Servers
1. AI Server for Enterprise Workloads(NVIDIA RTX Pro Server)
This is a modular GPU server that can be set up with approximately 8 high-performance NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. It supports AI implementation in service environments, simulation, and visual computing work, enabling consumers to choose the GPU, memory and storage setup according to their particular applications. It is developed for reasoning, AI application advancement, and the combination of AI into business systems.
2. AI Server for Large-Scale Training and Inference (NVIDIA HGX System)
This server is geared up with the NVIDIA HGX Rubin NVL8, a server board that adjoins 8 GPUs by means of NVIDIA NVLink, and embraces a Direct Liquid Cooling approach (* 3). It likewise supports NVIDIA ConnectX-9 SuperNIC adapters and NVIDIA BlueField-4 DPU setups, making it ideal for the training and reasoning of generative AI designs, consisting of Large Language Models (LLMs). Providing steady high-density, high-power operation, it is a design well matched for AI training facilities and core information center applications.
Typical to Models 1 and 2: High-Performance Networking
Both designs support cluster release based upon the NVIDIA Enterprise Reference Architecture (Enterprise RA), and by integrating NVIDIA Quantum InfiniBand and NVIDIA Spectrum-X Ethernet, they recognize the execution of scalable AI factories.
* 3 A cooling approach utilizing liquid rather of traditional air cooling to dissipate heat created inside devices more effectively.
Main Specifications of the AI Server for Enterprise Workloads (NVIDIA RTX Pro Server)
- Specs and look undergo alter without notification due to enhancements and other factors.
- Sharp is preparing to start accepting orders for the AI Server for Large-Scale Training and Inference (NVIDIA HGX System) in FY2027, and comprehensive requirements will be revealed when completed.
- Even within Japan, associated services might not be readily available sometimes due to geographical restraints, such as the range from upkeep service bases.
- NVIDIA, NVIDIA RTX Pro, NVIDIA Blackwell, NVIDIA HGX, NVIDIA Rubin, NVIDIA NVLink, NVIDIA ConnectX, NVIDIA BlueField, NVIDIA Quantum InfiniBand and NVIDIA Spectrum-X Ethernet are hallmarks and/or signed up hallmarks of NVIDIA Corporation in the United States and other nations.
- Intel and Xeon are hallmarks and/or signed up hallmarks of Intel Corporation or its subsidiaries in the United States and/or other nations.
Client Inquiries (Japanese just)
For assessment and questions, please check out the following site.
https://smj.jp.sharp/bs/ssl/sh/ai-server/form.php
To ask for a quote, please check out the following site.
https://smj.jp.sharp/bs/ssl/sh/ai-server_estimation/form.php
About Sharp
For more than 110 years, Sharp Corporation has actually been establishing pioneering, worldâEUR’very first and industry-first items and innovations mostly in electronic devices. Based upon its service creed “Sincerity and Creativity” the business has actually developed its business motto “In step with your future.” and intends to develop New Cultures through ingenious product or services in every element of how individuals live and work. For additional information, please check out: https://global.sharp/