ASUS XA NB3I-E12 10NVMe
- Condition
- New
- Manufacturer
- ASUS
- CPU
- 2x Intel Xeon 6 6767P (64C 336M Cache 2.40 GHz)
- RAM
- 32kh 128GB DDR5 RDIMM 6400MHz (Support up to 32 DIMM 4TB xrtov DDR5)
- Power supply unit
- 10x 3200W Titanium
- HDD
- 2x 1.92TB NVMe Gen5 PM9D3a
- NIC card
- 1x 2port x 200GbE ConnectX-7
- GPU
- 1 x NVIDIA HGX B300 8-GPU server
- Accelerator model
- NVIDIA B300 Blackwell Ultra (HGX B300)
- Installed GPU count
- 8
- Maximum GPU count
- 8
- Memory per GPU
- 288GB HBM3e
- Total GPU memory
- 2304GB
- GPU memory bandwidth
- 8000 GB/s (8.0 TB/s) per GPU
- GPU form factor
- SXM
- GPU interconnect
- Fifth-generation NVIDIA NVLink with NVSwitch
- GPU-to-CPU interface
- PCIe 5.0 through the HGX baseboard fabric
- Supported compute data types
- FP64, FP32, TF32, BF16, FP16, FP8, FP6, FP4, INT8
- GPU virtualization
- NVIDIA vGPU, SR-IOV, GPU passthrough
- MIG support
- Up to 7 isolated GPU instances per GPU
- Software ecosystem
- NVIDIA CUDA, NVIDIA AI Enterprise and NVIDIA Transformer Engine
- Cooling
- Air cooling; passive accelerator modules with dedicated chassis airflow and hot-swap fan modules
- Recommended workloads
- AI reasoning and inference at scale, large-model training, agentic and multimodal AI, HPC, simulation and data analytics
- Dimensions
- 394.5 x 447 x 945 mm (H x W x D)
- Weight
- 120 kg
Need a technical consultation on choosing a server?
Submit a requestASUS XA NB3I-E12 10NVMe is a high-density 9U enterprise GPU server storage configuration with ten front-accessible PCIe Gen5 NVMe hot-swap drive bays. It is designed for AI factories, research centers, cloud providers, and enterprise data centers that need fast local storage for large-scale AI training, inference, and HPC workloads.
Key features:
-
Storage basket and service access: The configuration provides ten front 2.5-inch hot-swap NVMe positions connected through the platform PCIe fabric. Front-accessible hot-swap serviceability helps reduce planned maintenance time.
-
Installed accelerator subsystem: The installed accelerator subsystem contains 8 (1 x 8-GPU HGX baseboard) NVIDIA B300 Blackwell Ultra SXM (HGX B300) accelerators with 2,304GB (8 x 288GB) of aggregate GPU memory. The configuration is sized for sustained enterprise workloads rather than a general-purpose display role.
-
High-bandwidth data path: The GPU fabric uses Fifth-generation NVIDIA NVLink with NVSwitch; 1.8 TB/s per GPU. Storage, GPU and network paths are arranged to support high-throughput data movement and minimize avoidable I/O bottlenecks.
-
Enterprise compute platform: The compute subsystem is configured as follows: CPU: 2 x Intel Xeon 6 6767P; memory: 4,096GB DDR5-6400; slots: 32 DDR5 DIMM; connectivity: embedded NVIDIA ConnectX-8 SuperNIC connectivity. This provides CPU, memory and PCIe resources for feeding the accelerator complex and hosting production services.
-
Management and availability: ASUS ASMB12-iKVM, ASUS Control Center and 5+5 x 3200W 80 PLUS Titanium redundant power supplies support remote administration, planned component replacement and resilient operation in the data center.
Use cases:
-
AI model training and fine-tuning: Uses the installed NVIDIA B300 Blackwell Ultra SXM (HGX B300) subsystem for deep learning, generative AI and large-model development.
-
Inference and production AI services: Provides GPU memory and parallel compute for LLM, multimodal and enterprise inference workloads.
-
HPC, simulation and analytics: Accelerates scientific computing, numerical simulation, data science and other parallel workloads supported by the NVIDIA B300 Blackwell Ultra SXM (HGX B300).
-
Data-intensive pipelines: Combines 10-bay PCIe Gen5 NVMe storage with high-bandwidth GPU and network paths for datasets, checkpoints, staging and local scratch storage.
ASUS XA NB3I-E12 10NVMe is a purpose-built 9U GPU platform for organizations that need 10-bay PCIe Gen5 NVMe storage, 2,304GB (8 x 288GB) of installed GPU memory and enterprise-class 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.