XENON NITRO GPU Servers
XENON NITRO GPU servers and systems are fully optimised for the latest GPU computing modules and deliver 10X higher application performance than the latest multi-core CPU systems, providing unprecedented real-world scalability and breakthrough power efficiency across a wide array of HPC and AI applications.
Configure Your System NowDelivering maximum acceleration for training, inference, HPC, and analytics.
- Single AMD EPYC™ series processor family (up to 280W)
- Up to 64-core, 128 threads
- Up to 2TB REG ECC DDR4-3200MHz; 8 DIMM slots
Delivering maximum acceleration for training, inference, HPC, and analytics.
- Dual AMD EPYC™ series processor family (up to 280W each)
- Up to 64-core, 128 threads per processor
- Up to 8TB REG ECC DDR4-3200MHz; 32 DIMM slots
The XENON NITRO™ GX49 Gen2 rack server supports up to 10x NVIDIA® GPU accelerators.
- Dual Intel® Xeon® Scalable Processors family
- Up to 8TB DDR4-3200 ECC Registered DIMM
- Support for a range of NVIDIA® GPUs including Quadro/Quadro RTX
The XENON NITRO™ G29 Gen2 rack server supports up to 6x NVIDIA® GPU accelerators.
- Dual Intel® Xeon® Scalable Processors family
- Up to 4TB DDR4-3200 ECC Registered DIMM
- Support for 6x Double Width Passive GPUs
- Supports a range of NVIDIA® including Quadro, Quadro RTX
Single Intel® Xeon® Scalable Processors in a 1RU form factor supporting up to 2 GPUs.
- Single CPU socket
- Support for Intel® Xeon® Scalable Processors
- Max. Core 28
- Up to 768GB ECC LRDIMM
- Up to DDR4-2666MHz; 6x DIMM slots
Supports up to 4 x NVIDIA® Tesla® A100 SXM4 GPUs and is ideal for Artificial Intelligence, Deep Learning and HPC workloads.
- Dual Intel® Xeon® Scalable Processors family
- Supports up to 8TB DDR4-3200 ECC Registered DIMM
- Up to 600 GB/s GPU-to-GPU
- NVIDIA® NVLINK™
- Optimised for NVIDIA® GPUDirect™ RDMA
XENON customised NVIDIA® GPU cluster solutions fit your precise use case or application needs. Contact us today to start designing your GPU Cluster solution.
- Build your cluster with customised servers and NVIDIA® Data Centre GPU cards or DGX A100 or DGX H100 based systems.
- Matched server architecture, complete with networking and data storage.
- Complete system racked, pre-configured and tested for your applications and workloads.