Supermicro SYS-821GE-TNHR 16NVMe+3SFF
- Condition
- New
- Manufacturer
- Supermicro
- Form factor
- 8U
- RAM
- 8192GB (DDR5 ECC REG)
- GPU
- 8 x NVIDIA H200
- Accelerator manufacturer
- NVIDIA
- Accelerator model
- NVIDIA H200
- Installed GPU/APU count
- 8
- Maximum GPU/APU count
- 8 onboard GPUs
- Memory per accelerator
- 141GB
- Total accelerator memory
- 1128GB
- GPU memory type
- HBM3e
- Accelerator memory bandwidth
- 4800 GB/s (4.8 TB/s) per GPU
- Accelerator form factor
- SXM5
- Accelerator interconnect
- Fourth-generation NVIDIA NVLink with NVSwitch
- Accelerator-to-CPU interface
- PCIe 5.0 x16 per GPU
- Supported compute data types
- FP64, FP32, TF32, BF16, FP16, FP8, INT8
- Accelerator virtualization
- NVIDIA vGPU, MIG-backed vGPU and GPU passthrough
- MIG support
- Up to 7 isolated GPU instances per GPU
- Software ecosystem
- NVIDIA CUDA, CUDA-X and NGC; NVIDIA AI Enterprise-compatible
- Maximum system power
- 9kW
- Cooling
- Air cooling with up to 10 heavy-duty fans and optimal fan speed control
- Recommended workloads
- Large-scale LLM training, fine-tuning and inference, generative AI, HPC, simulation, analytics and scientific computing
- Dimensions
- 355.6 x 437 x 843.28 mm (H x W x D)
- Weight
- 75.3 kg
Need a technical consultation on choosing a server?
Submit a requestSupermicro SYS-821GE-TNHR 16NVMe+3SFF is an 8U enterprise server storage configuration with sixteen front hot-swap 2.5-inch NVMe bays plus three front hot-swap 2.5-inch SATA bays. It is designed for AI factories, research centers, cloud providers and enterprise data centers that need fast local storage for AI, accelerated analytics and HPC workloads.
Key features:
-
Storage basket and service access: The selected storage basket comprises sixteen front hot-swap 2.5-inch NVMe bays plus three front hot-swap 2.5-inch SATA bays. Front- or rear-accessible hot-swap service, as applicable to the specified bays, helps reduce planned maintenance time without opening the main chassis.
-
Installed accelerator subsystem: The installed accelerator subsystem contains 8 x NVIDIA H200 SXM. It provides 1,128GB HBM3E (8 x 141GB). Fourth-generation NVIDIA NVLink with NVSwitch provides the high-bandwidth fabric inside the HGX H200 subsystem.
-
Storage topology and data path: The NVMe bays connect through the platform PCIe storage path. Storage, accelerator and network resources are arranged to support sustained data movement for datasets, checkpoints and local scratch space.
-
Enterprise compute platform: The compute subsystem combines dual 5th/4th Generation Intel Xeon Scalable processor platform with 8,192GB DDR5 ECC registered memory. It supplies the CPU, memory and PCIe resources required to feed the accelerator complex and host production services.
-
Management and availability: Dedicated BMC connectivity and Supermicro lifecycle-management tools, together with 6 x 3000W redundant (4+2) Titanium-level power supplies, support remote administration, component monitoring, planned service operations and resilient data-center operation.
Use cases:
-
AI model training and fine-tuning: Uses the installed NVIDIA H200 SXM subsystem for deep learning, generative AI and large-model development.
-
Inference and production AI services: Provides accelerator memory and parallel compute for LLM, multimodal and enterprise inference workloads.
-
HPC, simulation and analytics: Accelerates scientific computing, numerical simulation, engineering, analytics and other parallel workloads supported by the installed accelerator subsystem.
-
Data-intensive pipelines: Combines the specified 16NVMe+3SFF storage configuration with high-bandwidth accelerator and network paths for datasets, checkpoints, staging and local scratch storage.
Supermicro SYS-821GE-TNHR 16NVMe+3SFF is a purpose-built 8U platform for organizations that need the specified storage basket, installed NVIDIA H200 SXM acceleration and enterprise-class lifecycle management in one deployable system.
to Buy a Server
from Servermall
warranty
testing
We want to make sure that our equipment is fully reliable for you. That’s why all our equipment (even new) is subject to complete diagnostics, testing, and part replacement.
testing
We fully load our equipment with various utilities and software. Rest assured, it’ll work for a very long time.
Power supply
The components work at full capacity and are tested for several days. If we are not sure about something, we replace it.
of all equipment
Our regulations imply the same testing for both used and new equipment. Oh, and by the way, we complement the official manufacturer’s warranty with our own warranty.