本網站使用瀏覽器紀錄 (Cookies) 來提供您最好的使用體驗，我們使用的 Cookie 也包括了第三方 Cookie。相關資訊請訪問我們的隱私權與 Cookie 政策。如果您選擇繼續瀏覽或關閉這個提示，便表示您已接受我們的網站使用條款。 關閉
Retailers Can Leverage AI to Increase Revenue and Reduce Costs with Supermicro and NVIDIA's Collaboration SAN JOSE, Calif. and NEW YORK, Jan. 15, 2025 /PRNewswire/ -- National Retail Federation (NRF) -- Supermicro, Inc. (SMCI), a Total IT Solution Provider for AI/ML, HPC, Cloud, Storage, and 5G/Edge, is showcasing the latest solutions for the retail industry in collaboration with NVIDIA at the National Retail Federation (NRF) annual show. As generative AI (GenAI) grows in capability and becomes more easily accessible, retailers are leveraging NVIDIA NIM microservices, part of the NVIDIA AI Enterprise software platform, for a broad spectrum of applications. Super Micro AI Solutions for Enterprise and Retail "Supermicro's innovative server, storage, and edge computing solutions improve retail operations, store security, and operational efficiency," said Charles Liang, president and CEO of Supermicro. "At NRF, Supermicro is excited to introduce retailers to AI's transformative potential and to revolutionize the customer's experience. Our systems here will help resolve day-to-day concerns and elevate the overall buying experience." To learn more about Supermicro's AI solutions for retail organizations, visit www.supermicro.com/RetailAI Innovation in front-of-house and back-of-house use cases – such as personalized shopping experiences, automating logistical processes, and preventing shrinkage – enable retailers to attract customers to their establishments and improve their profitability. Supermicro at NRF Retail AI solutions like the ones mentioned above require specialized hardware to deliver optimal results. Supermicro offers the industry's broadest portfolio of edge AI solutions, giving retail organizations the tools to optimize the ROI of their AI-based applications. By processing data directly where the customers are located, results are delivered with minimal latency, creating an exceptional user experience. At NRF, Supermicro is showing the latest systems designed to deliver high-end performance to retail locations: SYS-112B-FWT: This 1U, short-depth system offers powerful edge compute, utilizing the latest Intel Xeon processors. Despite being less than 17" in depth, it can hold NVIDIA L40S GPUs. AS -1115S-FWTRT: Leveraging AMD's EPYC series processors, this 1U, short-depth system offers customers flexibility in both core counts and GPU capability by having the capacity to hold a single double-width NVIDIA L40S GPU. SYS-E403-14B: A compact server capable of delivering data center performance at remote and space-constrained locations. This compact system can accommodate 2 single-width or 1 double-width GPU cards, including the NVIDIA L40S or NVIDIA RTX 6000 Ada Generation GPU, in remote edge locations. SYS-212B-FN2T: The 2U, short-depth edge AI system offers more features and capacity compared to the competition and supports multiple data streams or workloads. Optimized for inferencing at the edge, the SYS-212B-FN2T supports up to 2 double-width GPU or single-width GPUs, such as the NVIDIA L4 GPU. SYS-222HE-TN: A 2U powerhouse, this dual-processor system features the latest Intel Xeon processors bringing data center power into a shorter-depth platform. With the ability to hold up to 3 NVIDIA L40S GPUs, it brings serious inferencing power to the customer. AS -2115HE-FTNR: A uniprocessor, 2U system based on the latest AMD EPYC family. This system offers a maximum dual-width GPU density of 4 cards such as NVIDIA L40S GPUs. SYS-322GA-NR: For the most demanding workloads, this powerful and versatile 3U edge AI platform offers 10 PCIe 5.0 x16 slots or 20 PCIe 5.0 x8 slots (in physical x16 slots). These slots can be fitted with up to 8 double-width GPU cards, such as NVIDIA's H100 GPU leveraging NVIDIA NVLink. Alternatively, the system can be configured with 19 single-width GPU cards, such as NVIDIA's low-profile or single-width product families, or combined with different add-on cards, such as graphics controllers, making this system ideal for environments including large control rooms. An AI use case on display at the Supermicro booth is a customer service application built with the NVIDIA AI Blueprint for digital humans. It features James, a computer interface that mimics the experience of interacting with a real person. Digital humans are virtual avatars that look, act, and think like a human would. In retail, this technology can be deployed to provide customers with a personalized shopping experience, a shopping assistant, and a customer service representative all in one solution. A second live demo is the NVIDIA AI Blueprint for retail shopping assistants, a generative AI reference workflow designed to transform shopping experiences online and in stores. The blueprint provides the technologies for building shopping assistants that can search for multiple items simultaneously, answer contextual questions like whether a product is waterproof, and visualize furniture in physically accurate scenes, such as a customer's living room. A third use case being demonstrated harnesses the NVIDIA AI Blueprint for video search and summarization to combat the growing issue of shrinkage. Shrinkage occurs when products leave a store and are not paid for. Loss due to shrinkage costs retailers an estimated $100+B annually in the US alone. By using AI-based solutions, retailers can interpret camera input in real time and identify mix-ups at the checkout as they occur, providing direct feedback to the retail store and the customer. NRF visitors can find Supermicro at booth #3165. At the NRF 2025 Expo, January 12-14 in New York City, Supermicro and NVIDIA are collaborating to introduce retailers to AI's transformative potential. For more information, visit https://nrfbigshow2025.smallworldlabs.com/co/super-micro-computer-inc 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 a Total IT Solutions provider with server, AI, storage, IoT, switch systems, software, and support services. Supermicro's motherboard, power, and chassis design expertise further enables our development and production, enabling next generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Asia, 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. All other brands, names, and trademarks are the property of their respective owners.
Completely Upgraded Servers with the New Intel® Xeon® 6900 Series Processors with P-Cores Optimized for Maximum Performance are Now Shipping in Volume, Featuring Next-Generation GPU Support, Higher-Bandwidth Memory, 400GbE Networking, E1.S and E3.S Drives, and the Industry's Leading Direct-to-Chip Liquid Cooling Solutions SAN JOSE, Calif., Jan. 9, 2025 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI) a Total IT Solution Provider for AI/ML, HPC, Cloud, Storage, and 5G/Edge is commencing shipments of max-performance servers featuring Intel Xeon 6900 series processors with P-cores. The new systems feature a range of new and upgraded technologies with new architectures optimized for the most demanding high-performance workloads including large-scale AI, cluster-scale HPC, and environments where a maximum number of GPUs are needed, such as collaborative design and media distribution. SBI-612BA-1NE34 "The systems now shipping in volume promise to unlock new capabilities and levels of performance for our customers around the world, featuring low latency, maximum I/O expansion providing high throughput with 256 performance cores per system, 12 memory channels per CPU with MRDIMM support, and high performance EDSFF storage options," said Charles Liang, president and CEO of Supermicro. "We are able to ship our complete range of servers with these new application-optimized technologies thanks to our Server Building Block Solutions® design methodology. With our global capacity to ship solutions at any scale, and in-house developed liquid cooling solutions providing unrivaled cooling efficiency, Supermicro is leading the industry into a new era of maximum performance computing." Click here for more information. There are currently several X14 systems available to customers for remote testing and validation via Supermicro's JumpStart program. The Supermicro X14 systems are available in a variety of form factors, each optimized for a range of performance-intensive workloads: GPU-optimized supporting the latest generation of SXM and PCIe GPUs, featuring enhanced thermal capacities and direct-to-chip liquid cooling on some models. High density multi-nodes including all-new FlexTwin™ and GrandTwin® models as well as the proven, award-winning SuperBlade® architecture. These models leverage shared components to increase efficiency and can be fitted with direct-to-chip liquid cooling for maximum performance density. Market-proven Supermicro Hyper rackmounts combine single or dual socket architectures with flexible I/O and storage configurations in traditional rackmount form factors to help enterprises and data centers scale up and out as their workloads evolve. Supermicro's max-performance X14 systems support the Intel Xeon 6900 series processors with P-cores, which feature up to 128 performance cores per CPU, support for high bandwidth MRDIMMs up to 8800 MT/s, and built-in accelerators including the AI-specific Intel AMX. The X14 systems represent the perfect building blocks for data centers at any scale, with Supermicro able to provide complete rack-level integration services including design, building, testing, validation, and delivery. An industry-leading global manufacturing capacity of up to 5,000 racks per month (2,000 liquid cooled) and extensive testing and burn-in facilities allow Supermicro to deliver solutions at any scale in a matter of weeks, not months. With Supermicro's complete in-house liquid-cooled direct-to-chip cold plate solutions, liquid cooling can be easily included in rack-level integrations to further increase system efficiency, reduce instances of thermal throttling, and lower both the TCO and Total Cost to Environment (TCE) of data center deployments. These turn-key solutions include the rack, cabling, power, and cooling infrastructure to simplify solution deployment at scale. To maximize the performance and density potential of the latest X14 systems, Supermicro also offers complete in-house developed liquid cooling solutions including cold plates for CPUs, GPUs, memory, Cooling Distribution Units, Cooling Distribution Manifolds, hoses, connectors, and cooling towers. Liquid cooling is easily included in rack-level integrations to increase system efficiency, reduce instances of thermal throttling, and lowers both the TCO and Total Cost to the Environment (TCE) of data center deployments. Supermicro max-performance X14 systems featuring Intel Xeon 6900 series processors with P-cores include: GPU-optimized – The highest performance Supermicro X14 systems designed for large-scale AI training, large language models (LLMs), generative AI and HPC, and supporting eight of the latest-generation SXM5 and SXM6 GPUs. These systems are available in air-cooled or liquid-cooled configurations. PCIe GPU – Designed for maximum GPU flexibility, supporting up to 10 double-width PCIe 5.0 accelerator cards in a thermally-optimized 5U chassis or edge-optimized 3U chassis. These servers are ideal for AI inferencing, media, collaborative design, simulation, cloud gaming, and virtualization workloads. Intel® Gaudi® 3 AI Accelerators – Supermicro is now shipping the industry's first AI server based on the Intel Gaudi 3 accelerator hosted by Intel Xeon 6 processors. Designed to increase the efficiency and lower the cost of large-scale AI model training and AI inferencing, the system features eight Intel Gaudi 3 accelerators on an OAM universal baseboard, six integrated OSFP ports for cost-effective scale-out networking, and an open platform designed to use a community-based, open-source software stack, requiring no software licensing costs. SuperBlade® – Supermicro's X14 6U high-performance, density-optimized, and energy-efficient SuperBlade maximizes rack density, with up to 100 servers and 200 GPUs per rack. Optimized for AI, HPC, and other compute-intensive workloads, each node features air cooling or direct-to-chip liquid cooling to maximize efficiency and achieve the lowest PUE with the best TCO, as well as connectivity up to four integrated Ethernet switches with 100G uplinks and front I/O supporting a range of flexible networking options up to 400G InfiniBand or 400G Ethernet per node. FlexTwin™ – The new Supermicro X14 FlexTwin architecture is purpose-built for HPC, cost-efficient, and designed to provide maximum compute power and density in a multi-node configuration with up to 24,576 performance cores in a 48U rack. Optimized for HPC and other compute-intensive workloads, each node features direct-to-chip liquid cooling only to maximize efficiency and reduce instances of CPU thermal throttling, as well as HPC low latency front and rear I/O supporting a range of flexible networking options up to 400G per node. Hyper – X14 Hyper is Supermicro's flagship rackmount platform designed to deliver the highest performance for demanding AI, HPC, and enterprise applications, with single or dual socket configurations supporting double-width PCIe GPUs for maximum workload acceleration. Both air cooling and direct-to-chip liquid cooling models are available to facilitate the support of top-bin CPUs without thermal limitations and reduce data center cooling costs while also increasing efficiency. 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 a Total IT Solutions provider with server, AI, storage, IoT, switch systems, software, and support services. Supermicro's motherboard, power, and chassis design expertise further enables our development and production, enabling next generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Asia, 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. All other brands, names, and trademarks are the property of their respective owners. SYS-212HA-TN SYS-212HA-TN-FRONT SYS-222FT-HEA-LCC-ANGLE SYS-222FT-HEA-LCC-FRONT SYS-422GA-NBRT-LCC SYS-522GA-NRT SYS-A22GA-NBRT
Supermicro的SuperCluster配備了NVIDIA HGXTM B200 8-GPU、NVIDIA GB200、NVL4和NVL72系統，提供空前的AI運算密度 加州聖荷西和亞特蘭大2024年11月22日 /美通社/ -- Supermicro, Inc. （NASDAQ：SMCI）為AI、雲端、儲存和5G/邊緣領域的全方位IT解決方案製造商，宣布推出最高效能的SuperCluster。此端對端AI資料中心解決方案採用NVIDIA Blackwell平台，專為兆級參數規模的生成式AI時代所設計。全新SuperCluster將可大幅增加液冷機架中的NVIDIA HGX B200 8-GPU系統數量，與Supermicro目前領先業界，搭載NVIDIA HGX H100和H200的液冷型SuperCluster相比，GPU運算密度得到大幅提升。此外，Supermicro正在擴大其NVIDIA Hopper系統產品組合，以因應加速型HPC應用和主流企業級AI技術的快速普及。 Supermicro總裁暨執行長梁見後表示：「Supermicro 擁有必要的專業技術、交付速度和產能，能部署包含100,000個GPU的全球最大液冷AI 資料中心專案，而透過Supermicro與NVIDIA的協作，該專案最近剛完成部署。這些Supermicro SuperCluster能透過高效直接液冷（DLC）技術降低電力需求。同時，我們現在推出了採用NVIDIA Blackwell平台的解決方案。透過我們的建構組件（Building Block）技術，我們能快速設計搭載NVIDIA HGX B200 8-GPU的伺服器，並能自由搭配液冷或氣冷散熱配置。我們的SuperCluster提供空前的密度、效能和效率，為未來更密集的AI運算解決方案奠定基礎。Supermicro的運算叢集搭配直接液冷技術，可幫助整個資料中心實現更高的效能和更低的功耗，並降低營運成本。」 欲了解更多資訊，請瀏覽：www.supermicro.com/hpc 經認可的大規模AI效能：Supermicro NVIDIA HGX B200系統 經升級後的SuperCluster可擴充單元採用機架級設計，具有創新垂直冷卻液分配歧管（CDM），可使單一機架內的運算節點數量增加。新開發的高效散熱冷板和先進的軟管設計可進一步改善液冷系統效率。適合大型部署的新型In-Row式冷卻液分配裝置（CDU）亦可被納入運算叢集的散熱配置內。傳統氣冷資料中心也能運用搭配新型氣冷系統機箱的全新NVIDIA HGX B200 8-GPU系統。 全新Supermicro NVIDIA HGX B200 8-GPU系統相較於前一代，提供了多項升級。這些新系統包含對散熱和電源配置的優化，並支援雙500W Intel® Xeon® 6（具備DDR5 MRDIMM，傳輸速度達8800 MT/s）或AMD EPYCTM 9005系列處理器。而全新一款氣冷式10U機型Supermicro NVIDIA HGX B200系統則採用經重新設計的機箱，具有更佳的散熱空間以支援8 個1000W TDP Blackwell GPU。這些系統在GPU搭配NIP的架構上採用1:1比例設計，並支援NVIDIA BlueField®-3 SuperNIC或NVIDIA ConnectX®-7 NIC，可在高效能運算結構內進行擴充。此外，每個系統所配備的兩個NVIDIA BlueField-3資料處理器（DPU），能使附加型高效能AI儲存裝置之間的資料傳輸處理效率得到提升。 Supermicro_NVIDIA_Blackwell 採用NVIDIA GB200 Grace Blackwell超級晶片的Supermicro解決方案 Supermicro亦提供搭配所有NVIDIA GB200 Grace Blackwell超級晶片的解決方案，包括最新發表的NVIDIA GB200 NVL4超級晶片和NVIDIA GB200 NVL72單機架百萬兆級電腦。 Supermicro的NVIDIA MGX系列設計將支援NVIDIA GB200 Grace Blackwell NVL4 超級晶片。這款超級晶片啟動了融合式高效能運算與AI的未來，並透過NVLink-C2C技術，以及四個由NVIDIA NVLink™ 連接的Blackwell GPU與兩個NVIDIA Grace™ CPU，提供了革命性的效能。這些超級晶片與Supermicro的液冷型NVIDIA MGX模組化系統相容，且相較於上一代產品，能為科學運算、圖神經網路（GNN）訓練和推論應用提供最高2倍的效能。 搭配Supermicro端對端液冷解決方案的NVIDIA GB200 NVL72 SuperCluster能在具有SuperCloud Composer（SCC）軟體的單個機架中組成一個百萬兆級超級電腦，為液冷資料中心提供全面的監控與管理能力。72個NVIDIA Blackwell GPU和36個NVIDIA Grace CPU皆透過第五代NVIDIA NVLink和NVLink Switch連接，能有效地作為單個強大GPU運行，並擁有高度容量的HBM3e記憶體集區，以實現低延遲的130TB/s總GPU通訊頻寬。 搭載NVIDIA H200 NVL的加速運算系統 Supermicro的 5U PCIe加速運算系統現可搭載NVIDIA H200 NVL，適用於需要彈性配置的低功耗、氣冷型企業級機架設計，並能為許多AI和高效能運算工作負載進行加速（無論機體尺寸大小）。NVIDIA H200 NVL能藉由NVIDIA NVLink，將最多四個GPU相連，並搭配1.5倍記憶體容量，以及透過HBM3e得到提高的1.2倍頻寬，可實現數小時內的LLM快速微調，同時也提供比前一代快1.7倍的LLM 推論效能速度。NVIDIA H200 NVL亦包含NVIDIA AI Enterprise的五年服務，而NVIDIA AI Enterprise是一個用於開發與部署生產型AI的雲端原生軟體平台。 Supermicro 的X14和H14 5U PCIe加速運算系統透過NVLink技術支援多達兩個4路NVIDIA H200 NVL系統，在單個系統內可搭載八個GPU，提供最高900GB/s的GPU互連傳輸速度，而每4-GPU NVLink域的組合集區具有 564GB 的 HBM3e 記憶體。全新PCIe 加速運算系統支援最高10個PCIe GPU，現在也可搭載最新Intel Xeon 6或AMD EPYC 9005系列處理器，能為高效能運算和AI應用提供彈性且多功能的選項。 Supermicro已參與2024年超級運算大會 Supermicro在超級運算大會上展示了完整的AI和高效能運算基礎架構解決方案組合，包括我們用於AI SuperCluster的液冷GPU伺服器。 透過我們展位的演講議程，能夠了解客戶、Supermicro專家及我們的技術合作夥伴所介紹的最新突破性運算技術。 歡迎蒞臨Supermicro在SC24展場B廳的2531號展位。 關於Super Micro Computer, Inc. Supermicro（納斯達克股票代碼：SMCI）為應用最佳化全方位IT解決方案的全球領導者。Supermicro的成立據點及營運中心位於美國加州聖荷西，致力為企業、雲端、AI和5G電信/邊緣IT基礎架構提供領先市場的創新技術。我們是全方位IT解決方案製造商，提供伺服器、AI、儲存、物聯網、交換器系統、軟體及支援服務。Supermicro的主機板、電源和機箱設計專業進一步推動了我們的發展與產品生產，為全球客戶實現了從雲端到邊緣的下一代創新。我們的產品皆由企業內部團隊設計及製造（在美國、亞洲及荷蘭），透過全球化營運實現極佳的規模與效率，並藉營運最佳化降低總體擁有成本（TCO），以及經由綠色運算技術減少環境衝擊。屢獲殊榮的Server Building Block Solutions®產品組合，讓客戶可以自由選擇這些具高度彈性、可重複使用且極為多元的建構式組合系統，我們支援各種外形尺寸、處理器、記憶體、GPU、儲存、網路、電源和散熱解決方案（空調、自然氣冷或液冷），因此能為客戶的工作負載與應用提供最佳的效能。 Supermicro、Server Building Block Solutions和We Keep IT Green皆為Super Micro Computer, Inc. 的商標和/或註冊商標。 所有其他品牌、名稱和商標皆為其各自所有者之財產。
Supermicro, Inc. （NASDAQ：SMCI）為AI、雲端、儲存和5G/邊緣領域的全方位IT解決方案製造商，宣布推出最高效能的SuperCluster。此端對端AI資料中心解決方案採用NVIDIA Blackwell平台，專為兆級參數規模的生成式AI時代所設計。全新SuperCluster將可大幅增加液冷機架中的NVIDIA HGX B200 8-GPU系統數量，與Supermicro目前領先業界，搭載NVIDIA HGX H100和H200的液冷型SuperCluster相比，GPU運算密度得到大幅提升。此外，Supermicro正在擴大其NVIDIA Hopper系統產品組合，以因應加速型HPC應用和主流企業級AI技術的快速普及。 Supermicro總裁暨執行長梁見後表示：「Supermicro 擁有必要的專業技術、交付速度和產能，能部署包含100,000個GPU的全球最大液冷AI 資料中心專案，而透過Supermicro與NVIDIA的協作，該專案最近剛完成部署。這些Supermicro SuperCluster能透過高效直接液冷（DLC）技術降低電力需求。同時，我們現在推出了採用NVIDIA Blackwell平台的解決方案。透過我們的建構組件（Building Block）技術，我們能快速設計搭載NVIDIA HGX B200 8-GPU的伺服器，並能自由搭配液冷或氣冷散熱配置。我們的SuperCluster提供空前的密度、效能和效率，為未來更密集的AI運算解決方案奠定基礎。Supermicro的運算叢集搭配直接液冷技術，可幫助整個資料中心實現更高的效能和更低的功耗，並降低營運成本。」 欲了解更多資訊，請瀏覽：www.supermicro.com/hpc 經認可的大規模AI效能：Supermicro NVIDIA HGX B200系統 經升級後的SuperCluster可擴充單元採用機架級設計，具有創新垂直冷卻液分配歧管（CDM），可使單一機架內的運算節點數量增加。新開發的高效散熱冷板和先進的軟管設計可進一步改善液冷系統效率。適合大型部署的新型In-Row式冷卻液分配裝置（CDU）亦可被納入運算叢集的散熱配置內。傳統氣冷資料中心也能運用搭配新型氣冷系統機箱的全新NVIDIA HGX B200 8-GPU系統。 全新Supermicro NVIDIA HGX B200 8-GPU系統相較於前一代，提供了多項升級。這些新系統包含對散熱和電源配置的優化，並支援雙500W Intel® Xeon® 6（具備DDR5 MRDIMM，傳輸速度達8800 MT/s）或AMD EPYCTM 9005系列處理器。而全新一款氣冷式10U機型Supermicro NVIDIA HGX B200系統則採用經重新設計的機箱，具有更佳的散熱空間以支援8 個1000W TDP Blackwell GPU。這些系統在GPU搭配NIP的架構上採用1:1比例設計，並支援NVIDIA BlueField®-3 SuperNIC或NVIDIA ConnectX®-7 NIC，可在高效能運算結構內進行擴充。此外，每個系統所配備的兩個NVIDIA BlueField-3資料處理器（DPU），能使附加型高效能AI儲存裝置之間的資料傳輸處理效率得到提升。 採用NVIDIA GB200 Grace Blackwell超級晶片的Supermicro解決方案 Supermicro亦提供搭配所有NVIDIA GB200 Grace Blackwell超級晶片的解決方案，包括最新發表的NVIDIA GB200 NVL4超級晶片和NVIDIA GB200 NVL72單機架百萬兆級電腦。 Supermicro的NVIDIA MGX系列設計將支援NVIDIA GB200 Grace Blackwell NVL4 超級晶片。這款超級晶片啟動了融合式高效能運算與AI的未來，並透過NVLink-C2C技術，以及四個由NVIDIA NVLink™ 連接的Blackwell GPU與兩個NVIDIA Grace™ CPU，提供了革命性的效能。這些超級晶片與Supermicro的液冷型NVIDIA MGX模組化系統相容，且相較於上一代產品，能為科學運算、圖神經網路（GNN）訓練和推論應用提供最高2倍的效能。 搭配Supermicro端對端液冷解決方案的NVIDIA GB200 NVL72 SuperCluster能在具有SuperCloud Composer（SCC）軟體的單個機架中組成一個百萬兆級超級電腦，為液冷資料中心提供全面的監控與管理能力。72個NVIDIA Blackwell GPU和36個NVIDIA Grace CPU皆透過第五代NVIDIA NVLink和NVLink Switch連接，能有效地作為單個強大GPU運行，並擁有高度容量的HBM3e記憶體集區，以實現低延遲的130TB/s總GPU通訊頻寬。 搭載NVIDIA H200 NVL的加速運算系統 Supermicro的 5U PCIe加速運算系統現可搭載NVIDIA H200 NVL，適用於需要彈性配置的低功耗、氣冷型企業級機架設計，並能為許多AI和高效能運算工作負載進行加速（無論機體尺寸大小）。NVIDIA H200 NVL能藉由NVIDIA NVLink，將最多四個GPU相連，並搭配1.5倍記憶體容量，以及透過HBM3e得到提高的1.2倍頻寬，可實現數小時內的LLM快速微調，同時也提供比前一代快1.7倍的LLM 推論效能速度。NVIDIA H200 NVL亦包含NVIDIA AI Enterprise的五年服務，而NVIDIA AI Enterprise是一個用於開發與部署生產型AI的雲端原生軟體平台。 Supermicro 的X14和H14 5U PCIe加速運算系統透過NVLink技術支援多達兩個4路NVIDIA H200 NVL系統，在單個系統內可搭載八個GPU，提供最高900GB/s的GPU互連傳輸速度，而每4-GPU NVLink域的組合集區具有 564GB 的 HBM3e 記憶體。全新PCIe 加速運算系統支援最高10個PCIe GPU，現在也可搭載最新Intel Xeon 6或AMD EPYC 9005系列處理器，能為高效能運算和AI應用提供彈性且多功能的選項。 Supermicro已參與2024年超級運算大會 Supermicro在超級運算大會上展示了完整的AI和高效能運算基礎架構解決方案組合，包括我們用於AI SuperCluster的液冷GPU伺服器。 透過我們展位的演講議程，能夠了解客戶、Supermicro專家及我們的技術合作夥伴所介紹的最新突破性運算技術。 歡迎蒞臨Supermicro在SC24展場B廳的2531號展位。
全新的 FlexTwin™ 和新一代 SuperBlade®，每個機架最高可達 36,864 核心，具有直接晶片液體冷卻和升級技術，以最大幅度提高 HPC 效能 加州聖何塞和亞特蘭大2024年11月21日 /美通社/ -- Supercomputing Conference -- Supermicro, Inc.（納斯達克：SMCI）為 AI/ML、HPC、雲端、儲存和 5G/Edge 提供全面 IT 解決方案供應商，展示其最新的高運算密度多節點解決方案，已就高強度 HPC 工作負載進行最佳化。這些系統包括創新的液體冷卻 FlexTwin 2U 4 節點專用 HPC 架構，以及業界領先的 SuperBlade SuperBlade，可在 6U 或 8U 機箱中最多 20 個節點，並提供多種儲存磁碟機選項。每個 SuperBlade 節點可容納一個 NVIDIA GPU，加速特定應用程式。Supermicro 多重節點以共用資源為特色，可提高效率並降低原物料的使用量，與標準機架安裝系統相比，密度大幅改善。 4U-B200-sys Supermicro 主席暨行政總裁梁見後 (Charles Liang) 表示：「自 2007 年推出業界第一部 Twin 系統以來，Supermicro 一直在開發最密集、最高效率的多節點架構方面的先驅。全新 FlexTwin 包含帶 P 核心的 Intel Xeon 6 處理器，或是全新的 AMD EPYC 9005 處理器，讓客戶可以選擇 HPC 機架規模部署。結合 Supermicro 液體冷卻經驗、廣泛的多節點開發專業知識、機架規模整合能力和最新的產業技術，使我們能夠為客戶提供具有前所未有的性能和規模的 HPC 解決方案，協助解決全球最複雜的運算挑戰。」 請瀏覽 www.supermicro.com/hpc，查看這些新系統，並進一步了解 Supermicro 在 SC24 展覽會上的展出情況。 Supermicro 多節點系統已針對 HPC 工作負載進行最佳化，包括金融服務、製造業、氣候與天氣建模、石油與天然氣，以及科學研究。每個獨特的產品系列，都是以密度、效能和效率的最佳組合而設計。 FlexTwin — 全新的雙處理器平台，專為液體冷卻多節點架構中提供最高效能密度而設計，並支援最新的 CPU、記憶體、儲存和冷卻技術。FlexTwin 專為大規模支援嚴苛的 HPC 工作負載（包括金融服務、製造、科學研究和複雜建模）而設計，性能出色且性價比高。使用 48U 機架作為例，FlexTwin 在此機架尺寸內最多可支援 96 個雙處理器節點和 36,864 個核心。 SuperBlade — 高效能、密度最佳化且節能的架構，每個機架最多可容納 100 台伺服器和 200 個 GPU。直接液體冷卻 (DLC) 能夠支援使用最高功率 CPU 的伺服器，以最佳的總擁有成本 (TCO) 達到最低的電力使用效率 (PUE)。SuperBlade 利用共用的備援元件，包括電源供應器、冷卻風扇、機箱管理模組 (CMM)、乙太網路和 InfiniBand 交換器，以及通過模組，提供最符合成本效益的綠色運算解決方案。根據客戶需求，靈活的 SuperBlade 提供 6U 或 8U 外型規格。 BigTwin — 多功能的 Supermicro BigTwin 可提供 2U-4Node 或 2U-2Node 系統。超微型 BigTwin 共用電源供應器和風扇，可降低功耗。BigTwin 可搭配 Intel Xeon 6 處理器。 憑藉其完整機架整合服務，Supermicro 與客戶密切合作，為 HPC 工作負載設計和設計機架和整個資料中心解決方案。經過客戶密切參與的設計驗證後，Supermicro 提供現場部署服務，縮短部署時間。Supermicro 的產業足跡遍佈全球，在美國、歐洲和亞洲設有生產設施。公司每月總共可生產 5,000 個機架，包括 2,000 個液體冷卻機架，交貨時間只需數星期，而不是數個月。 Supermicro 的多節點系統採用最新技術，以提高 HPC 效能。 新一代處理器 — 具有高達 500 瓦的 P 核心和 128 核心的雙 Intel® Xeon 6900 系列處理器，或 AMD EPYC™ 9005 系列高達 500 瓦和 192 核心的處理器，可在一系列 Supermicro HPC 伺服器中使用。此外，雙 Intel Xeon 6700 處理器的 E 核心最高可達 330 瓦和 144 核心，也可在特定的 Supermicro 伺服器中使用。 更高頻寬記憶體 — 支援高達 6400MT/s DDR5，可提升記憶體密集型和記憶體內運算 HPC 應用程式的輸送量。採用 Intel Xeon 6 處理器的系統也支援頻寬高達 8800MT/s 的全新 MRDIMM。 直接液體冷卻 — Supermicro 的完整液體冷卻解決方案包括 CPU 和 DIMM 模組冷板、冷卻分配管 (CDM)、機架內和列式冷卻配電單元 (CDM)、連接器、軟管和冷卻塔，可有效率地冷卻高功率 CPU 並減少熱節流的情況。Supermicro 在過去三個月內部署了超過 2,000 個完全整合式液體冷卻機架。 EDSFF 磁碟機 — EDSFF E1.S 和 E3.S 硬碟機的全新支援，有助提高儲存密度並增強吞吐量，為資料密集的 HPC 應用提供更佳的儲存效能。EDSFF 磁碟機也具有比標準儲存磁碟機更有效率的散熱設計，因此在空間限制的多節點架構中可以提高磁碟機密度。 Supermicro 參加 Supercomputing Conference 2024 Supermicro 將在 Supercomputing Conference 展覽上展示完整的 AI 和 HPC 基礎架構解決方案產品組合，包括我們用於 AI 超級叢集 (AI SuperClusters) 的液冷 GPU 伺服器。 在我們的攤位影院中，客戶、Supermicro 的專家以及技術合作拍檔將發表演講，介紹運算科技的最新突破。 在 SC24 展位 B 展廳 #2531 參觀 Supermicro 的展出。 Super Micro Computer, Inc. 簡介 Supermicro（納斯達克股票代碼：SMCI）是應用優化整體 IT 解決方案的全球領軍企業。Supermicro 在加利福尼亞州聖荷西成立並營運，致力於為企業、雲端、AI 和 5G 電信/Edge IT 基礎設施提供率先進入市場的創新。我們是全面 IT 解決方案製造商，提供伺服器、AI、儲存、物聯網、交換器系統、軟件和支援服務。在主機板、電源供應器和機箱設計方面的專業知識，進一步支援 Supermicro 的研發和生產，使我們能夠為全球客戶提供從雲端到邊緣運算的下一代創新解決方案。我們內部（在美國、亞洲和荷蘭）設計和製造產品，利用全球營運來實現規模和效率，務求提高整體擁有成本 (TCO) 和減少對環境的影響（綠色運算）。獲獎無數的 Server Building Block Solutions® 透過我們靈活可重用的構建模組，為客戶提供各適其適的系統產品系列，用於優化其確切的工作負載和應用程式。這些構建模組支援全系列外形規格、處理器、記憶體、GPU、儲存、網絡、電源和冷卻解決方案（空調、自然空氣冷卻或液體冷卻）。 Supermicro、Server Building Block Solutions 和 We Keep IT Green 是 Super Micro Computer, Inc. 的商標和/或註冊商標。 所有其他品牌、名稱和商標均為其各自所有者的財產。 10U-HGX-B200 SYS-522GA-NRT
All-new FlexTwin™ and New Generation SuperBlade® Maximize Compute Density with up to 36,864 Cores per Rack, featuring Direct to Chip Liquid Cooling and Upgraded Technologies to Maximize HPC Performance SAN JOSE, Calif. and ATLANTA, Nov. 19, 2024 /PRNewswire/ -- Supercomputing Conference -- Supermicro, Inc. (NASDAQ: SMCI) a Total IT Solution Provider for AI/ML, HPC, Cloud, Storage, and 5G/Edge, is showcasing its latest high-compute-density multi-node solutions which are optimized for high intensity HPC workloads. These systems include the innovative liquid-cooled FlexTwin 2U 4-node purpose-built HPC architecture and the industry leading SuperBlade with up to 20 nodes in a 6U or an 8U chassis with a range of storage drive options. Each SuperBlade can accommodate an NVIDIA GPU per node, accelerating specific applications. Supermicro multi-nodes feature shared resources to enhance efficiency and reduce raw materials usage, with significant density improvements compared to standard rackmount systems. 4U-B200-sys "Since the release of the industry-first Twin systems back in 2007, Supermicro has consistently been a pioneer in developing the densest, most efficient multi-node architectures for HPC workloads," said Charles Liang, president and CEO of Supermicro. "The new FlexTwin houses either the Intel Xeon 6 processors with P-cores or the new AMD EPYC 9005 processors, giving customers the choice for their HPC racks scale deployments. The combination of Supermicro's liquid cooling experience, extensive multi-node development expertise, rack-scale integration capacity, and the latest industry technologies enables us to provide our customers with HPC solutions of unprecedented performance and scale, helping to solve the world's most complex computational challenges." See these new systems and learn more about Supermicro at SC24 by visiting www.supermicro.com/hpc. The Supermicro multi-node systems are optimized for HPC workloads, including financial services, manufacturing, climate & weather modeling, oil & gas, and scientific research. Each distinct product family is designed with an optimized combination of density, performance, and efficiency. FlexTwin – All-new dual-processor platform designed for maximum performance density in a liquid-cooled multi-node architecture, featuring support for the latest CPUs, memory, storage, and cooling technologies. Purpose-built to support demanding HPC workloads at scale, including financial services, manufacturing, scientific research, and complex modeling, FlexTwin is optimized for performance per dollar. Using a 48U rack as an example, FlexTwin can support up to 96 dual processor nodes and 36,864 cores within this rack size. SuperBlade - High-performance, density-optimized, and energy-efficient architecture with up to 100 servers and 200 GPUs per rack. Direct liquid cooling (DLC) can support servers with the highest power CPUs to achieve the lowest PUE with the best TCO. SuperBlade utilizes shared, redundant components, including power supplies, cooling fans, chassis management modules (CMMs), ethernet and InfiniBand switches, and pass-thru modules to deliver the most cost-effective, green computing solutions. Depending on customer requirements, the flexible SuperBlade is available in either a 6U or 8U form factor. BigTwin – The versatile Supermicro BigTwin is available as a 2U-4Node or 2U-2Node system. The Supermicro BigTwin shares power supplies and fans, which reduce power consumption. The BigTwin is available with the Intel Xeon 6 processor. With its Complete Rack Integration Services, Supermicro works closely with customers to architect and design rack and entire data center solutions for HPC workloads. After the design is validated with customer close involvement, Supermicro offers on-site deployment services, reducing the time-to-deployment. Supermicro has a global manufacturing footprint with production facilities in the USA, Europe, and Asia. It can produce a total of 5,000 racks per month, including 2,000 liquid-cooled racks, with lead times of weeks, not months. Supermicro's multi-node systems feature the latest technologies to enhance HPC performance. New-generation processors – Dual Intel® Xeon 6900 series processors with P-cores up to 500W and 128 cores or AMD EPYC™ 9005 series processors up to 500W and 192 cores are available in a range of Supermicro HPC servers. In addition, dual Intel Xeon 6700 processors with E-cores up to 330W and 144 cores are also available in select Supermicro servers. Higher bandwidth memory - Support for DDR5 up to 6400MT/s improves throughput for memory-intensive and in-memory computing HPC applications. Systems featuring Intel Xeon 6 processors also support the new MRDIMMs with a bandwidth of up to 8800MT/s. Direct Liquid Cooling – Supermicro's complete liquid cooling solutions include CPU and DIMM module cold plates, cooling distribution manifolds (CDM), in-rack and in-row cooling distribution units (CDM), connectors, tubing, and cooling towers, efficiently cooling high-powered CPUs and reducing instances of thermal throttling. Supermicro has deployed more than 2,000 fully integrated liquid-cooled racks in the past three months. EDSFF drives – New support for EDSFF E1.S and E3.S drives improves storage density and enhances throughput, providing better storage performance for data-intensive HPC applications. EDSFF drives also have a more efficient thermal design than standard storage drives, allowing a higher drive density in space-constrained multi-node architectures. Supermicro at Supercomputing Conference 2024 Supermicro will showcase a full AI and HPC infrastructure solutions portfolio at the Supercomputing Conference, including our liquid-cooled GPU servers for AI SuperClusters. Check out the speaking sessions at our in-booth theater, where customers, experts from Supermicro, and our technology partners will be presenting on the latest breakthroughs in computing technology. Visit Supermicro at booth #2531, Hall B at SC24. 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 a Total IT Solutions manufacturer with server, AI, storage, IoT, switch systems, software, and support services. Supermicro's motherboard, power, and chassis design expertise further enable our development and production, enabling next-generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Asia, 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. All other brands, names, and trademarks are the property of their respective owners. 10U-HGX-B200 SYS-522GA-NRT
A12 藝術空間
Supermicro
請先登入後才能發佈新聞。
還不是會員嗎？立即 加入台灣產經新聞網會員 ，使用免費新聞發佈服務。 （服務項目） （投稿規範）
