Showing 550–560 of 695 results

The Supermicro SYS-420GU-TNXR is a high-density 4U rackmount server engineered for intensive AI, machine learning, and high-performance computing (HPC) workloads. Built on the Supermicro Universal GPU platform, this system supports dual 5th or 4th Gen Intel Xeon Scalable processors, providing a robust foundation for data-intensive applications. Its primary differentiator is the integration of the NVIDIA HGX H100 or H200 4-GPU baseboard, utilizing high-speed NVLink interconnects to bypass traditional PCIe bottlenecks for massive parallel processing tasks.

Supermicro GPU server AS-4124GQ-TNMI is with direct-attached backplane supporting 10 hot-plug NVMe or SATA drives. Supports AMD Instinct MI250 OAM Accelerators with Infinity Fabric Link GPU-GPU Interconnect and which have 4 or 8 dedicated network cards. On board dedicated RJ45 port for IPMI 2.0 remote management with KVM and remote virtual media mount.

Supermicro SYS-421GE-TNHR2-LCC is a 4U GPU SuperServer designed for extreme AI, HPC, and deep learning workloads. Powered by 5th/4th Generation Intel Xeon Scalable processors, it supports up to eight NVIDIA HGX H100 or H200 GPUs with a liquid-cooling solution for optimal thermal performance. The server features a direct-attached backplane accommodating eight hot-plug SATA or NVMe drives, delivering high-speed storage performance. A dedicated RJ45 port with IPMI 2.0 ensures secure remote management and KVM access for easy monitoring. Engineered for next-generation data centers, this platform offers exceptional compute density, GPU scalability, and energy-efficient operation for demanding enterprise and research applications.

The Supermicro AS-4125GS-TNHR2-LCC is a high-density GPU A+ Server designed for AI, deep learning, HPC, and large-scale analytics. Powered by AMD EPYC 9004 Genoa processors and supporting up to eight NVIDIA HGX H100 or H200 GPUs with a liquid-cooling solution, it delivers unmatched computational performance for demanding workloads. The server includes a direct-attached backplane with 8 hot-plug NVMe drives for ultra-fast storage access and high I/O throughput. A dedicated RJ45 IPMI 2.0 port enables secure remote management, KVM access, and virtual media mounting. Engineered for enterprise and research data centers, the AS-4125GS-TNHR2-LCC provides scalable, energy-efficient, and reliable GPU-accelerated computing for next-generation applications.

The Supermicro SYS-2028TP-DTTR is a high-density 2U TwinPro™ system engineered for enterprise data centers requiring maximum compute density and power efficiency. This multi-node architecture features four independent hot-plug nodes housed in a single chassis, with each node supporting dual Intel Xeon E5-2600 v3/v4 processors. It is specifically designed for high-availability environments, hyper-converged infrastructure (HCI), and cloud service providers where rack space optimization is a primary concern.

Supermicro AS-4126GS-NMR-LCC-G1 4U GPU A+ Server is engineered for advanced AI training, HPC workloads, and large-scale data processing. Powered by dual AMD EPYC 9575F processors (64 cores each, 3.3GHz) and paired with 2.3TB of high-speed DDR5-6400 memory, it delivers exceptional compute performance. The system integrates eight AMD Instinct MI325X GPUs, offering powerful parallel acceleration for demanding AI and scientific applications. Storage includes a 480GB M.2 NVMe SSD with additional front PCIe 5.0 NVMe drive bays for expansion. Connectivity features eight 400GbE QSFP112 ports and dual 10GbE RJ45. Redundant 6600W Titanium power supplies ensure maximum reliability. Ideal for high-density GPU clusters requiring efficiency, scalability, and enterprise-grade performance.

The Supermicro SYS-424GP-TNAR+ is a high-density 4U rackmount server engineered specifically for large-scale AI training, Large Language Model (LLM) development, and complex High-Performance Computing (HPC) workloads. Built around the NVIDIA HGX H100 8-GPU baseboard, this system leverages high-speed NVLink and NVSwitch interconnects to provide massive GPU-to-GPU bandwidth, which is essential for distributed deep learning tasks and reducing synchronization bottlenecks.

The Supermicro SYS-421GE-TNRT is a high-density 4U rackmount server engineered for intensive AI/ML training, deep learning, and high-performance computing (HPC) workloads. Built on the 4th Gen Intel Xeon Scalable processor architecture, this system leverages the performance of dual Sapphire Rapids CPUs to handle complex data processing tasks. Its primary differentiator is the support for up to eight double-width PCIe 5.0 x16 GPUs, providing massive parallel computing power for neural network training and large-scale simulations.

The Supermicro SYS-421GE-TNRT3 is a high-density 4U rackmount server engineered for intensive GPU-accelerated workloads. Built on the 4th Gen Intel Xeon Scalable processor architecture (Sapphire Rapids), this system supports up to 8 double-width PCIe 5.0 x16 GPUs, making it a cornerstone for deep learning, molecular modeling, and high-performance computing (HPC) environments. It addresses the critical bottleneck of data throughput by utilizing PCIe 5.0 lanes, ensuring maximum bandwidth between the CPUs and hardware accelerators.

Supermicro AS-4145GH-TNBR is a high-performance GPU A+ Server designed for AI, HPC, and large-scale analytics workloads. It supports quad AMD Instinct MI300A accelerators, delivering massive parallel processing power for deep learning, scientific computing, and data-intensive applications. The server features a direct-attached backplane supporting up to 24 drives, with 8 hot-plug NVMe drives included by default, while SATA and SAS drives are supported via additional components. A dedicated RJ45 IPMI 2.0 port allows secure remote management, KVM access, and virtual media mounting. Engineered for energy-efficient, scalable, and reliable operation, the AS-4145GH-TNBR is ideal for research labs, AI infrastructure, and enterprise HPC deployments.