MiTAC Computing Unveils Advanced AI Cluster and Cooling Solutions at Supercomputing 2025
Mon, 17 Nov 2025 22:00:00 +0800
ST. LOUIS, Nov. 17, 2025 /PRNewswire/ -- MiTAC Computing Technology, a global leader in high-performance and energy-efficient server solutions, is proud to announce its participation at the Supercomputing (SC) 2025 (November 18–20, St. Louis, Missouri), booth 3916. Under the theme "AI Cluster Power – Cool Fast Scale Faster" MiTAC will showcase its modular highly scalable rack infrastructure for demanding AI and HPC workloads demonstrating capabilities from single servers to cluster integration focusing on liquid cooling and energy-efficient designs. MiTAC collaborates with AMD, Broadcom, CoolIT, Intel, KIOXIA, Micron, NVIDIA, Samsung and Solidigm to accelerate advanced computing and high-efficiency data centers.
Rack-scale Innovation From Standard Architecture to AI Cluster Excellence
At SC 2025 MiTAC Computing presents its full range of rack-level solutions for cluster-scale deployments including liquid-cooled and air-cooled AI and HPC racks addressing both open architectures and traditional enterprise data centers.
AI Liquid-Cooled Rack with AMD Instinct™ MI355X GPUs | Optimized for Ultra-Scale AI
The high-density 48U EIA AI liquid-cooled rack MR1100 series targets hyperscale AI training and inference. This robust solution supports 64 to 256 AI GPUs. Utilizing cold-plate technology with the latest AMD Instinct™ MI355X GPUs, AMD EPYC™ 9005 CPUs, and AMD Pensando™ Pollara 400 AI NICs, it ensures non-throttled AI throughput with 400/800 Gb/s network architecture.
AI Air-Cooled Rack with AMD Instinct™ MI350X GPUs | Standardized Architecture with High-Speed Interconnect
MiTAC presents its standard EIA 45U air-cooled AI rack, MR1100A, with four MiTAC G8825Z5 systems using AMD Instinct™ MI350X/MI325X GPUs. The 800Gb/s network switch Broadcom Tomahawk 5 chipset ensures low-latency data transmission GC68C-B8056 management and TS70A-B8056 storage servers enable rapid deployment of scalable AI/HPC clusters.
OCP ORv3 Liquid-Cooled Rack | Modular Power and Advanced Thermal Management
MiTAC's 43OU OCP ORv3 liquid-cooled rack, MR is designed for sustainable HPC, accommodating up to 14 C2811Z5 multi-node servers powered by AMD EPYC™ 9005 Series processors. It integrates MiTAC Lake Erie storage, a 33kW Power Shelf, and the CoolIT 200kW CHx200+ In-Rack CDU. This modular design ensures stable, energy-efficient operation for high-density compute servers.
AI Acceleration Platforms
- G4527G6: This 4U AI Accelerator, built on the NVIDIA MGX™ architecture, hosts up to 8 NVIDIA RTX PRO™ 6000 Blackwell Server Edition GPUs. It is driven by dual Intel® Xeon® 6767P CPUs with Solidigm D7-P5520 SSDs, providing the speed essential for deep learning.
- G4826Z5: A liquid-cooled compute node for MiTAC AI clusters, supporting up to eight AMD Instinct™ MI355X GPUs. It features full liquid cooling and includes the Broadcom P2200G network adapter.
- G8825Z5: A High-performance AI platform supporting up up to 8 AMD Instinct™ MI325X or MI350X GPUs. It also features AMD Pensando™ Pollara 400 AI NICs and Micron 6550 ION NVMe SSDs vital for accelerating training cycles.
HPC & Cloud Computing Frameworks
- G4520G6: This highly adaptable server, powered by the latest Intel® Xeon® 6 CPUs, delivers superior performance per watt for cloud and HPC roles, leveraging Micron DDR5 DRAM. It also features extensive expansion options, including support for NVIDIA RTX PROTM 6000 Blackwell Server Edition and NVIDIA Hopper GPUs.
- C2811Z5: An OCP-compliant liquid-cooled multi-node server, it uses AMD EPYC™ 9005 series processors. Optimized with NVMe E1.S interfaces and Micron 9550 NVMe SSDs, it guarantees sustained high-speed data delivery for intensive HPC.
Enterprise Data Solutions
- M2810Z5: Optimized for peak storage performance by deploying Kioxia XD8 E1.S Gen5 SSDs, this server unlocks maximum bandwidth for IO-intensive database applications.
- R2520G6: A high-capacity 2U server with dual Intel® Xeon® 6700P series processors integrates Solidigm D7-PS1010 SSDs. PCIe 5.0 bandwidth guarantees stability for data warehousing and analytics.
- M2510G6: This high-reliability platform supports Samsung MZTL67T6HBLC-00AW7 SSDs for mission-critical enterprise settings and demanding data processing.
- R2513G6: This platform provides maximum archival capacity using Seagate EXOS M 30TB HDDs, tailored for organizations with immense data retention needs.
Global Success Stories Demonstrating MiTAC's Value in AI & HPC Deployments Worldwide
At SC 2025, MiTAC showcases real-world success cases highlighting its proven expertise in delivering comprehensive rack-level solutions and its commitment to cluster-scale integration.
- MiTAC and French CSP pioneer sustainable HPC achieving world-class PUE and operational cost reduction: Partnered with Qarnot using the Capri 3 OCP server to capture 95% of server heat for reuse, achieving a PUE of 1.01 and reducing operational costs by 50% for clients in Europe. The exhibit also highlights the integration of the Broadcom N1400GD network adapter.
- OCP deployment accelerated in the US data centers featuring ORv3 platform and resource efficiency: MiTAC partnered with CTCA, a global IT solutions provider, to accelerate OCP deployment across the US data centers. This success leveraged the ORv3-compliant OCP server and advanced rack integration. The SC showcase highlights the OCP server featuring Samsung M321R8GA0EB2-CCP Memory, emphasizing faster rollouts and resource efficiency.
- Rack integration streamlines global operations for leading cloud security provider: MiTAC provided customized, ready-to-deploy rack integration services, utilizing the GC68C-B8056 server. This service streamlined operations by delivering over 380 configurations across 150 data centers worldwide, meeting 48-hour turnaround times, and enabling a GPU-first infrastructure shift within half a year, demonstrating the consistency, agility, and innovation MiTAC provides.
- MiTAC and its software-defined storage partner maximize GPU performance by eliminating AI training bottlenecks: Collaborating with a leading software-defined storage (SDS) partner, MiTAC leveraged the GC68A-B8056 server to remove critical I/O constraints. The pre-validated architecture boosted GPU utilization by 35% and delivered over 200 GB/s throughput, effectively tripling the speed of AI training completion.
For more SC information and product catalogs please visit:
About MiTAC Computing Technology Corporation
MiTAC Computing Technology Corp., a subsidiary of MiTAC Holdings, delivers comprehensive, energy-efficient server solutions backed by industry expertise dating back to the 1990s. Specializing in AI, HPC, cloud, and edge computing, MiTAC Computing employs rigorous methodologies to ensure uncompromising quality—across barebones, systems, racks, and cluster levels—fully achieving performance and integration. This commitment to quality at every level sets MiTAC Computing apart in the industry.
With a worldwide presence and end-to-end capabilities—from R&D and manufacturing to global support—MiTAC Computing provides agile, customized platforms for hyperscale data centers, HPC, and AI applications, ensuring optimal performance and scalability to meet unique business needs. By leveraging the latest advancements in AI and liquid cooling, and unifying the MiTAC brand with Intel DSG and TYAN server products, MiTAC Computing stands out for its innovative, efficient, and reliable server technology as well as its hardware and software integrated solutions—empowering businesses to meet future challenges.
MiTAC Computing Technology Corporation website: https://www.mitaccomputing.com/
Homepage
