Home > News > featured news > Supermicro Launches Industry's First NVIDIA HGX H100 8 and 4-GPU H100 Servers with Liquid Cooling -- Reduces Data Center Power Costs by Up to 40%

Hyper-Converged Infrastructure, Storage Management, IT Systems Management

Supermicro Launches Industry's First NVIDIA HGX H100 8 and 4-GPU H100 Servers with Liquid Cooling -- Reduces Data Center Power Costs by Up to 40%

Prnewswire | May 22, 2023 | Read time : 06:00 min

Supermicro Launches Industry&amp;#39;s First NVIDIA

Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for Cloud, AI/ML, Storage, and 5G/Edge, continues to expand its data center offerings with liquid cooled NVIDIA HGX H100 rack scale solutions. Advanced liquid cooling technologies entirely from Supermicro reduce the lead time for a complete installation, increase performance, and result in lower operating expenses while significantly reducing the PUE of data centers. Savings for a data center are estimated to be 40% for power when using Supermicro liquid cooling solutions compared to an air-cooled data center. In addition, up to 86% reduction in direct cooling costs compared to existing data centers may be realized.

"Supermicro continues to lead the industry supporting the demanding needs of AI workloads and modern data centers worldwide," said Charles Liang, president, and CEO of Supermicro. "Our innovative GPU servers that use our liquid cooling technology significantly lower the power requirements of data centers. With the amount of power required to enable today's rapidly evolving large scale AI models, optimizing TCO and the Total Cost to Environment (TCE) is crucial to data center operators. We have proven expertise in designing and building entire racks of high-performance servers. These GPU systems are designed from the ground up for rack scale integration with liquid cooling to provide superior performance, efficiency, and ease of deployments, allowing us to meet our customers' requirements with a short lead time."

To learn more about Supermicro's GPU servers, visit: https://www.supermicro.com/en/products/gpu

AI-optimized racks with the latest Supermicro product families, including the Intel and AMD server product lines, can be quickly delivered from standard engineering templates or easily customized based on the user's unique requirements. Supermicro continues to offer the industry's broadest product line with the highest-performing servers and storage systems to tackle complex compute-intensive projects. Rack scale integrated solutions give customers the confidence and ability to plug the racks in, connect to the network and become more productive sooner than managing the technology themselves.

The top-of-the-line liquid cooled GPU server contains dual Intel or AMD CPUs and eight or four interconnected NVIDIA HGX H100 Tensor Core GPUs. Using liquid cooling reduces the power consumption of data centers by up to 40%, resulting in lower operating costs. In addition, both systems significantly surpass the previous generation of NVIDIA HGX GPU equipped systems, providing up to 30x performance and efficiency in today's large transformer models with faster GPU-GPU interconnect speed and PCIe 5.0 based networking and storage.

State-of-the-art eight NVIDIA H100 SXM5 Tensor Core GPU servers from Supermicro for today's largest scale AI models include:

SYS-821GE-TNHR – (Dual 4th Gen Intel Xeon Scalable CPUs, NVIDIA HGX H100 8 GPUs, 8U)
https://www.supermicro.com/en/products/system/GPU/8U/SYS-821GE-TNHR
AS -8125GS-TNHR – (Dual 4th Gen AMD EPYC CPUs, NVIDIA HGX H100 8 GPUs, 8U)
https://www.supermicro.com/en/products/system/GPU/8U/AS-8125GS-TNHR

Supermicro also designs a range of GPU servers customizable for fast AI training, vast volume AI inferencing, or AI-fused HPC workloads, including the systems with four NVIDIA H100 SXM5 Tensor Core GPUs.

SYS-421GU-TNXR - (Dual 4th Gen Intel Xeon Scalable CPUs, NVIDIA HGX H100 4 GPUs, 4U)
https://www.supermicro.com/en/products/system/GPU/4U/SYS-421GU-TNXR
SYS-521GU-TNXR - (Dual 4th Gen Intel Xeon Scalable CPUs, NVIDIA HGX H100 4 GPUs, 5U)
https://www.supermicro.com/en/products/system/GPU/4U/SYS-521GU-TNXR

Supermicro's liquid cooling rack level solution includes a Coolant Distribution Unit (CDU) that provides up to 80kW of direct-to-chip (D2C) cooling for today's highest TDP CPUs and GPUs for a wide range of Supermicro servers. The redundant and hot-swappable power supply and liquid cooling pumps ensure that the servers will be continuously cooled, even with a power supply or pump failure. The leak-proof connectors give customers the added confidence of uninterrupted liquid cooling for all systems. Learn more about the Supermicro Liquid Cooling system at: https://www.supermicro.com/en/solutions/liquid-cooling

Rack scale design and integration has become a critical service for systems suppliers. As AI and HPC have become an increasingly critical technology within organizations, configurations from the server level to the entire data center must be optimized and configured for maximum performance. The Supermicro system and rack scale experts work closely with customers to explore the requirements and have the knowledge and manufacturing abilities to deliver significant numbers of racks to customers worldwide.

Read the Supermicro Large Scale AI Solution Brief - https://www.supermicro.com/solutions/Solution-Brief_Rack_Scale_AI.pdf

Supermicro at ISC

To explore these technologies and meet with our experts, plan on visiting the Supermicro Booth D405 at ISC High Performance 2023 event in Hamburg, Germany, May 21 – 25, 2023.

About Super Micro Computer, Inc.

Supermicro (NASDAQ: SMCI) is a global leader in Application-Optimized Total IT Solutions. Founded and operating in San Jose, California, Supermicro is committed to delivering first to market innovation for Enterprise, Cloud, AI, and 5G Telco/Edge IT Infrastructure. We are transforming into a Total IT Solutions provider with server, AI, storage, IoT, and switch systems, software, and services while delivering advanced high-volume motherboard, power, and chassis products. The products are designed and manufactured in-house (in the US, Taiwan, and the Netherlands), leveraging global operations for scale and efficiency and optimized to improve TCO and reduce environmental impact (Green Computing). The award-winning portfolio of Server Building Block Solutions® allows customers to optimize for their exact workload and application by selecting from a broad family of systems built from our flexible and reusable building blocks that support a comprehensive set of form factors, processors, memory, GPUs, storage, networking, power and cooling solutions (air-conditioned, free air cooling or liquid cooling).

Supermicro, Server Building Block Solutions, and We Keep IT Green are trademarks and/or registered trademarks of Super Micro Computer, Inc.

Spotlight

E-BOOK Umgestaltung der Telekommunikation mit offener Innovation

Bringen Sie Agilitat und Flexibilität in Ihre Systeme und Prozesse, damit Sie sich am Markt differenzieren und einen Wettbewerbsvorteil verschaffen. Finden Sie heraus, wie Serviceanbieter mit Red Hat ihre Kompetenzen und technische Kultur modernisieren können, urn ihre transformativen Möglichkeiten auszuschöpfen und die Open Hyb

More featured news

Spotlight

E-BOOK Umgestaltung der Telekommunikation mit offener Innovation

Related News

Application Infrastructure, Storage Management

Scala Data Centers Invests US$ 80 Million in the First Hyperscale Data Center in Tepotzotlán, Mexico

prnewswire | August 24, 2023

Scala Data Centers, the leading Latin American platform of sustainable data centers in the Hyperscale market, is building its first data center in Mexico, SMEXTP01, with an initial investment of approximately US$ 80 million. The venture, located in the municipality of Tepotzotlán, about 25 miles from the country's capital in Mexico City (CDMX), will have an initial critical capacity of 5MW reinforcing Scala's strategy of enabling new markets in Latin America to meet the massive demands being made by large cloud and content providers, known as Hyperscale. Scala's new site is projected to begin operations in Q4 2023 and is already anchored by a large Hyperscale client. With 142,133 sqft (about 13.2 thousand square meters) of built area, SMEXTP01 HyperEdge data center will offer maximum customization and modularity and will be supplied with 100% renewable and certified energy, in line with the company's portfolio. "This new project underscores our commitment to enabling new markets in Latin America, either in unexplored locations or in regions with data center presence, but still lacking a dedicated approach to large IT capacities that truly serve the Hyperscale segment," explains Marcos Peigo, CEO and Co-Founder at Scala. "We are proud to spearhead this expansion and create unique conditions for massive colocation clients in the region through innovation, energy efficiency, and sustainable technology," he adds. Geographically, Mexico has a relevant and strategic position in the IT infrastructure segment. As the second largest data center and cloud market in Latin America, the country has robust growth rates of around 22% and 25% per year, respectively, and excellent connectivity infrastructure. Currently, there are 11 submarine cables connecting Mexico to the United States and the rest of the Americas. Alongside Chile, where Scala also operates and invests, Mexico is one of the only countries in the region that will have a direct connection via submarine cable to the Asian market. Scala's SMEXTP01 in Tepotzotlán city will uniquely bring together an ideal scenario for the hyperscale market - energy availability, connectivity, and proximity to a large consumer market for critical IT infrastructure. In addition to the investment in SMEXTP01, Scala continues its accelerated expansion plan, with the construction of new sites at Tamboré Campus (São Paulo, Brazil), Porto Alegre city (Rio Grande do Sul state, Brazil), Lampa (Chile), as well as advanced projects to implement other data centers in Fortaleza city (Ceará state, Brazil), Mexico and Colombia. Since the company's inception in 2020, Scala has invested more than R$ 8 billion (US$ 1,6 billion) in its projects in Latin America and expects to reach 500MW of distributed capacity over the next five years, in four countries in which it operates. In 2023, the company started operating three state-of-the-art data centers in Latin America, located in Rio de Janeiro (Brazil), (13.2 MW of IT), Tamboré (Brazil) with 6MW of IT and Curauma (Chile) offering 5MW of IT. About Scala Data Centers Scala Data Centers is the leading Latin American platform of sustainable data centers in the Hyperscale market. Headquartered in Brazil and founded by DigitalBridge, it was developed to meet and exceed the growing demand for digital access in Latin America. Scala has a highly qualified team of over 800 professionals and applies a flexible and innovative approach to providing exceptional quality colocation services to hyperscale clients, cloud-based software and service providers, and large enterprises. We customize state-of-the-art solutions for each client in the construction of the latest generation data centers, with high availability, the best energy efficiency rates, and superior density. All this allied to the best sustainability practices guided by our ESG (Environmental, Social, Governance) program.

Application Infrastructure

Fermaca Networks Partners with Gold Data to provide Networks Services to US and Mexican Customers

Businesswire | July 06, 2023

Fermaca Networks (FN), a purpose built Infrastructure provider seeking to utilize the most unique right of way for purposes of being a leading provider of Long Haul Dark Fiber infrastructure in Mexico, announces its Partnership with Gold Data that will allow both companies to leverage complementary strengths in the marketplace. In addition, Gold Data is acquiring an undisclosed number of Fiber Pairs in Fermaca Networks phase 1 route (FN-1) which originates in El Paso, Texas and traverses multiple cities in Mexico and reaches as far South as Queretaro and Guadalajara. “FN’s core strengths complement those of Gold Data perfectly and we are excited about the opportunity to align our values,” stated Edgar Mosti, CRO, Fermaca Networks. “The skillsets and experiences of the two companies work synergistically and clients originating from either partner can expect a smooth and successful joint solution for fiber requirements throughout the areas they serve. We are sharply focused on our clients’ success, and we believe this partnership will benefit our mutual and prospective customers.” We are excited to also welcome Gold Data into the Fermaca Fiber Partners (FFP) program in which Gold Data will play a key role. We are excited to acquire fiber pairs in Fermaca's network, Gold Data complements its state-of-the-art topology network that is revolutionizing Mexico's demanding connectivity market. We are supporting this market with an offer from several international networks. "With this new underground route, it allows us to increase Gold Data's reliable footprint in addition to our 'Gold Data 1' backbone submarine cable. Also, we are very happy to join Fermaca Fiber Partners, who will provide extended networks to all Fermaca customers," says Renato Tradardi, CEO, Gold Data Group. About Gold Data Gold Data is a multinational award-wining technology company, which focuses on telecommunications infrastructure and services. We provide direct connectivity through the Americas and the Caribbean, interconnecting more than 156 PoPs, 76 data centers with presence in 35 countries in Latin America and the Caribbean. Gold Data relies on its strategic PoPs, diverse subsea capacity and its fully-owned network to deliver the complete end-to-end solution to and from Latin America and the Caribbean for MNCs, global carriers, OTT and content provider customers. About Fermaca With roots going back 60 years, Fermaca Dreams is a conglomerate of companies focused on developing and executing key infrastructure projects in Mexico. With proven track record, a successful history of partnerships, and a strong and professional management team, Fermaca Dreams companies are committed to the highest standards in project development, execution and operation.

Windows Server OS

Kentik Becomes the First Network Observability Platform to Completely Map Traffic and Performance across Hybrid and Public Clouds

PRnewswire | June 29, 2023

Kentik, the network observability company, today announced general availability of Kentik Map for Azure and extensive support for Microsoft Azure infrastructure in Kentik Cloud.With this release, Kentik provides comprehensive network observability across all multi-cloud deployments, and gives network, cloud, and infrastructure engineers the context to rapidly answer any question about their networks. Enterprises use Kentik Cloud to optimize performance, resolve network incidents, reduce cloud costs, and maximize ROI of migrations to Azure, Google Cloud, and AWS. These latest Azure enhancements follow the recently announced release of Kentik Map for Google Cloud, underscoring Kentik's command of network observability for hybrid and multi-cloud environments. Kentik Map for Azure visualizes Azure infrastructure in an interactive, data- and context-rich map that highlights how Azure resources nest within each other and connect to on-prem environments. Additionally, Kentik now collects, analyzes, and contextualizes traffic flow and performance data from Microsoft Azure along with data from on-prem, Google Cloud, and AWS services, enriching all network telemetry with deep application, business, and security context to provide actionable observability across all hybrid and multi-cloud environments. Organizations can now: • Reference live visualizations of Azure infrastructure topology in Kentik Map for Azure, with both current and historical views of paths and detailed metadata • Quickly understand path and performance details of traffic routes that connect Microsoft Azure workloads to data centers across Azure Express Routes • Visualize and interrogate traffic across Azure Firewalls, Load Balancers, Application Gateways, and other network infrastructure • Visualize and interrogate all branch, VPN, private cloud, and intra-cloud connectivity that is connected with Azure VWAN • Drill into the impact of NSG (Network Security Group) configuration on security and productivity with live traffic data • Use custom tags to add any custom context to Microsoft Azure data along with pre-built Azure-native attributes sourced from Azure APIs • Create custom dashboards with configurable pre-built widgets to visualize important Azure cost and performance trends "Legacy NPM and observability tools are simply not able to provide the visibility that enterprises need to understand the impact of traffic flows and performance on customer experiences," said Christoph Pfister, Chief Product Officer, Kentik. "By unifying all network telemetry, across all clouds and data centers, Kentik's latest advancements enable enterprises to make data-driven decisions, optimize resource utilization, and achieve peak performance and security for their hybrid and multi-cloud infrastructure." About Kentik Kentik is the leading observability company across the network, infrastructure, and cloud. Our platform is a must-have for the network front line, whether digital business, corporate IT or service provider. Network professionals turn to Kentik to plan, run and fix any network, relying on our infinite granularity, AI-driven insights and fast search. Kentik makes sense of network, infrastructure, cloud, host and container flow, internet routing, performance tests and network metrics. We show IT what they need to know about their network performance, health and security to make their business-critical services shine. Market leaders like Akamai, Booking.com, Dropbox, and Zoom rely on Kentik for network observability.