GPU Servers

AMAX GPU servers provide unprecedented real-world scalability and breakthrough power-efficiency across a wide array of Deep Learning and HPC applications. Our highly configurable GPU systems benefit from unparalleled performance that enables the seamless integration of GPU computing modules, helping scientists and engineers solve the world's most complex computational problems.

ServMax G140

2x Xeon Scalable Processors

Features

• 1U DP Xeon® Scalable Family
  Processors 4x Tesla GPU Server.
• Delivers up to 2,560 Tensor cores,
  20,480 CUDA Cores, 28 Teraflops DP,
  56 Teraflops SP, and 448 Teraflops
  Tensor performance.

×

ServMax G240

2x Xeon Scalable Processors

Features

• 2U DP Xeon Xeon® Scalable Family
  Processors 4x Tesla GPU Server.
• Delivers up to 2,560 Tensor cores,
  20,480 CUDA Cores, 28 Teraflops double
  precision, 56 Teraflops single
  precision, and 448 Teraflops
  Tensor performance
  in a 2U chassis.
• Ideal for HPC, deep learning,
  artificial intelligence,
  datacenter applications

×

ServMax G280

2x Xeon Scalable Processors

Features

• 2U DP Xeon® Scalable
  Family Processors 8x
  Tesla GPU Server.
• Delivers up to 5,120
  Tensor cores, 40,960
  CUDA Cores, 56 Teraflops DP,
  112 Teraflops SP, and 896
  Teraflops Tensor performance.
• Ideal for HPC, deep learning,
  artificial intelligence,
  datacenter applications

×

DL-E48A

2x Xeon Scalable Processors

Features

• 4U DP Xeon® Scalable Family Processors
  8x Tesla GPU Server
• Up to 5,120 Tensor cores, 40,960 CUDA
  Cores, 56 Teraflops double
  precision, 112 Teraflops single precision,
  and 896 Teraflops Tensor performance.
• Flexible GPU utilization with
  single & dual root complex design
• Ideal for HPC, deep learning,
  artificial intelligence,
  datacenter applications

×

DL-E48G

2x Xeon Scalable Processors

Features

• 4U DP Xeon® Scalable Family Processors
  8x Tesla PCIe GPU Server
• Up to 5,120 Tensor cores, 40,960 CUDA
  Cores, 56 Teraflops double
  precision, 112 Teraflops single precision,
  and 896 Teraflops Tensor performance.
• Flexible GPU utilization with
  single complex design
• Ideal for HPC, deep learning,
  artificial intelligence,
  datacenter applications

×

DL-E48S

2x Xeon Scalable Processors

Features

• 4U DP Xeon® Scalable Family Processors
  8x Tesla SXM2 GPU Server
• Up to 5,120 Tensor cores, 40,960 CUDA
  Cores, 56 Teraflops double
  precision, 112 Teraflops single precision,
  and 896 Teraflops Tensor performance.
• Ideal for HPC, deep learning,
  artificial intelligence,
  datacenter applications

×

ServMax G480

2x Xeon Scalable Processors

Features

• 4U DP Xeon® Scalable Family
  Processors 8x
  Tesla GPU Server
• Up to 5,120 Tensor cores,
  40,960 CUDA Cores,
  56 Teraflops double precision,
  112 Teraflops single precision,
  and 896 Teraflops Tensor performance.
• Ideal for HPC, deep learning,
  artificial intelligence,
  datacenter applications

×

Model	FF	CPU	GPU	Storage	Networking	REF #
ServMax G140	1U	2x Intel® Xeon® Processor Scalable Family	4x NVIDIA Tesla V100 SXM2 / PCIe	2x 2.5" hot swap + 2x 2.5" int	2x 1G, 2x 10G optional	Q706427 (1GbE)
ServMax G240	2U	2x Intel® Xeon® Processor Scalable Family	4x NVIDIA Tesla V100 PCIe	8 x H. S 3.5"	2x 1G, 2x 10G optional	Q706286 (1GbE)
ServMax G280	2U	2x Intel® Xeon® Processor Scalable Family	8x NVIDIA Tesla V100 PCIe	8x 2.5" hot swap + 2x 2.5"NVMe	2x 10G	Q706434
ServMax DL-E48A	4U	2x Intel® Xeon® Processor Scalable Family	8x NVIDIA Tesla V100 PCIe	8x 2.5" hot-swap, 2x M.2p	2x 1G (option 1) 2x 1G, 10G opt., IB FDR opt.	Q705986
ServMax DL-E48G	4U	2x Intel® Xeon® Processor Scalable Family	8x NVIDIA Tesla V100 PCIe	10x 2.5" hot-swap + 12x 3.5" hot-swap	2x 10G	Q707586
ServMax DL-E48S	4U	2x Intel® Xeon® Processor Scalable Family	8x NVIDIA Tesla V100 SXM2	10x 2.5" hot-swap	2x 1G 2x 10G optional	Q707587
ServMax G480	4U	2x Intel® Xeon® Processor Scalable Family	8x NVIDIA Tesla V100 SXM2 / PCIe	24x 2.5" hot-swap	2x 1G 2x 10G optional	Q706436