4U 2-Socket AMD EPYC Accelerator Server with 8 PCIe 4.0 GPUs supported
- AMD EPYC 7003 and 7003 with AMD 3D V-Cache Series Processors
- Supports up to 8 x PCIe Gen 4.0 dual-slot GPU cards, and up to another 3 x PCIe Gen4 slots for PCIe network adapters or other expansion cards.
- Optional OCP3.0 NIC socket support for networking flexibility
- Capacity for 8 x 3.5“ / 2.5’’ hot-swap storage bays, and the backplane could be configured to support 8 x NVMe drives. (Additional NVMe RAID card is needed.)
Reference # Q716811
The AceleMax DGS-428A is a a powerful GPU architecture that supports up to eight dual-slot active or passive GPUs, including NVIDIA® A100 and AMD Instinct™ MI100, with an optimized internal thermal layout. With NVIDIA NVLink® Bridge or AMD Infinity Fabric™ Link to enable scaling of performance, AceleMax DGS-428A lets you meet the demands of AI and high-performance computing.
Powered by AMD EPYC™ 7003 series processors
The AMD EPYC™ 7003 processor features the world’s first 7-nanometer data-center design and a PCI Express® 4.0-capable x86 server processor with up to 2X the performance and 4X the floating-point capability. With support for up to 64 cores and 128 threads, 8-channel DDR4 3200, 128 PCIe 4.0 lanes and up to a 280-watt maximum thermal design power (TDP), this processor also includes memory and virtualization encryption features to ensure exceptional security.
Target markets: AI / Deep Learning, Cloud Computing, Virtualized & VDI Application, and Enterprise & HPC Application.
Built with AMD EPYC™ 7003 Series Processor
Providing incredible compute, IO and bandwidth capability – designed to meet the huge demand for more compute in big data analytics, HPC and cloud computing.
- Built on 7nm advanced process technology, allowing for denser compute capabilities with lower power consumption
- Up to 64 core per CPU, built using Zen 2 high performance cores and AMD’s innovative chiplet architecture
- Supporting PCIe Gen 4.0 with a bandwidth of up to 64GB/s, twice of PCIe Gen 3.0
- Embedded security protection to help defend your CPU, applications, and data
NVIDIA A100 PCIe GPU
The NVIDIA A100 Tensor Core GPU delivers unprecedented acceleration and flexibility to power the world’s highest-performing elastic data centers for AI, data analytics and HPC applications. As the engine of the NVIDIA data center platform, the A100 GPU provides up to 20X higher performance and 2.5X AI performance than V100 GPUs, and can efficiently scale up to thousands of GPUs or be partitioned into seven isolated GPU instances with new multi-Instance GPU (MIG) capability to accelerate workloads of all sizes.
The NVIDIA A100 GPU features third-generation Tensor Core technology that supports a broad range of math precisions providing a unified workload accelerator for data analytics, AI training, AI inference, and HPC. It also supports new features such as New Multi-Instance GPU, delivering optimal utilization with right sized GPU and 7x Simultaneous Instances per GPU; New Sparsity Acceleration, harnessing Sparsity in AI Models with 2x AI Performance; 3rd Generation NVLINK and NVSWITCH, delivering Efficient Scaling to Enable Super GPU, and 2X More Bandwidth than the V100 GPU. Accelerating both scale-up and scale-out workloads on one platform enables elastic data centers that can dynamically adjust to shifting application workload demands. This simultaneously boosts throughput and drives down the cost of data centers.
S Y S T E M
P R O C E S S O R S
2 x Socket SP3 (LGA-4094)
Supports AMD EPYC™ 7002 & 7003 series processors (up to 280W)
G P U
8 x PCIe 4.0 x16 link for dual-slot GPU cards
M E M O R Y
Up to 32 DIMM slots, Up to 4TB
D R I V E S
8 x SATA/SAS/NVMe Devices
I / O
Front I/O Ports: 2 x USB 3.2 Gen1 ports, 1 x VGA port
Rear I/O Ports: 1 x Management port (RJ45)
C O O L I N G F A N S
5 heavy duty fans
P O W E R S U P P L I E S
2+2 Redundant 2200W 80 PLUS Platinum Power Supply or
2+2 Redundant 2200W 80 PLUS Titanium Power Supply or
2+2 Redundant 3000W 80 PLUS Titanium Power Supply
- 2 x Socket SP3 (LGA-4094)
- Supports AMD EPYC™ 7003/7002 Series with up to 64 cores, 128 threads, and 256MB L3 cache per processor, or EPYC™ 7003 with AMD 3D V-Cache Series, with up to 64 cores, 128 threads, and 768MB L3 cache per processor
- Up to 280W TDP
- 32 (8-channel per CPU) slots
- Up to 4TB
- DDR4 3200 RDIMM/LRDIMM memory type
Graphics Processing Unit (GPU):
- 8 x PCIe Gen 4.0 dual-slot GPU cards, and up to another 3 x PCIe Gen4 slots for PCIe network adapters or other expansion cards
- 8 x PCIe x16 slots (Gen4 x16 link, FH,FL) for dual-slot GPU cards
- 1 x PCIe x16 slot (Gen4 x16/x8 link, FH,HL) or OCP3.0 NIC socket (optional)
- 1 x PCIe x8 slot (Gen4 x8 link, LP,HL) (optional)
- Up to 24 x 2.5″ SAS/SATA drive bays
- 4 x 2.5″ SATA supported natively
- 4 x 2.5″ NVMe supported natively
- 1 x dedicated management port (BMC AST2600)
- 2+2 Redundant 2200W 80 PLUS Platinum Power Supply or
- 2+2 Redundant 2200W 80 PLUS Titanium Power Supply or 2+2
- Redundant 3000W 80 PLUS Titanium Power Supply
- 800mm x 439mm x 175.6mm (4U)
- 31.4” x 17.22” x 6.92”