S5G | D52G-4U

Need to do 560 Teraflops in a Single Rack..?The D52G-4U also known as D5G is the world's fastest  GPU server. Hyperconverged: 8 * NVMe with up to 16 * GP-GPU the D5G does 56 Double Precision Teraflops in 4U. It is i

Need to do 560 Teraflops in a Single Rack..?

The D52G-4U also known as D5G is the world's fastest  GPU server. Hyperconverged: 8 * NVMe with up to 16 * GP-GPU the D5G does 56 Double Precision Teraflops in 4U. It is ideal for: Big Data, Deep learning, Artificial Intelligence and Inferencing. D52G boasts: NVLink™ support up to 300GB/s GPU to GPU communication | 10x dual-width 300 Watt GPU | 16x single-width 75 Watt GPU support | Diversify GPU topology to Conquer Any Type of Parallel Computing Workload | Up to 4x100Gb/s High Bandwidth RDMA-enabled Network to Scale Out with Efficiency | 8x NVMe Storage to Accelerate Deep Learning.



Deep learning with up to 16 GPU featuring with NvlinkTM

With NVLinkTM to support up to 300GB/s GPU to GPU communication, D52G-4U can shorten model-training time-frame, which is 40x faster than conventional server on processing 408GB Chest X-ray images.



10x dual-width 300 watt GPU or 16x single-width 75 Watt GPU support

As an purpose-built system for Artificial Intelligent (AI) and High-Performance Computing(HPC) workloads, QuantaGrid D52G-4U can deliver up to 896 tensor Tflops to training deep learning model with eight* NVIDIA® Tesla V100 dual-width 10.5 inch or provides up to 293 GOPS/watt of peak INT8 performance to do inferencing with sixteen NVIDIA® Tesla P4 and 2-socket Intel® Xeon® Scalable processor; up to 56 double precision Tflops computing power can accelerate HPC workloads such as Oil & Gas, bioinformatics, Mechanical Engineering. On top of superior computing power, D52G is  of  2x100Gb/s high bandwidth low-latency networking to expedite communication among  different GPU nodes



Diversify GPU Topology to Conquer Any Type of Parallel Computing Workload

The QuantaGrid D52G provides multiple GPU topology on the same baseboard tray to meeting your different use case.


High-Bandwidth, Low-latency Networking Between GPU Nodes

The QuantaGrid D52G-4U have additional two/four PCIe Gen3x16  LP-MD2 slot to provide optional two/four 100Gb/s low-latency infiniband or Intel®Omni Path options to do  GPUDirect in the server or RDMA between different GPU nodes.


NVMe SSD Support to Accelerate Deep Learning

The D52G-4U support max. 8x NVMe SSD, which can accelerate both training and inferencing with Fast I/O data reading as deep learning is a data-driven algorithm.


Processor
Processor TypeIntel®Xeon® Processor Scalable Family
Max. TDP Support165W
Number of Processors2 Processors
Internal Interconnect9.6/10.4 GT/s
Form Factor
Form Factor4U
Dimensions
W x H x D (inch)17.5 x 7x 34.6
W x H x D (mm)444x 177.8x 880
Chipset
ChipsetIntel® C621
Storage
Default Configuration
NVMe support
2.5" Hot-plug
SKU - #1(24) 2.5" hot-plug SATA/SAS SSD
or (20) 2.5" hot-plug SATA/SAS SSD + (4) 2.5" hot-plug NVMe
or (8) 2.5" hot-plug NVMe
Memory
Total Slots24
CapacityUp to 3TB (128Gx24) of memory for RDIMM/LRDIMM
Memory Type2666 MHz DDR4 RDIMM
Memory Size
64GB, 32GB, 16GB, 8GB RDIMM
(Refer to CCL)
Expansion Slot
Expansion Slot_Motherboard(1) PCIe Gen3 x 8 OCP 2.0 mezzanine slot (can switch to U.2)
(1) PCIe Gen3 x 8 optional SAS mezzanine slot (can switch to U.2)
(1) PCIe Gen3 x 8 FHHL(can switch to U.2) + (1) PCIe Gen3 x 8 LP-MD2 (can switch to U.2)
or optional (1) PCIe Gen3 x16 LP-MD2
Expansion Slot_GPGPU baseboard[Option 1]
(8) PCIe Gen3 x 16, SXM2 slot
(4) PCIe Gen3 x16, LP-MD2
[Option 2]
(8) PCIe Gen3 x16, dual-width FHFL + (2) PCIe Gen3 x16, LP-MD2 
or (10)PCIe Gen3 x16 dual-width FHFL
[Option 3]
(20) PCIe Gen3 x16, single-width FHFL
or (16) PCIe Gen3 x16, single-width FHFL + (4) PCIe Gen3 x16, LP-MD2
Network Controller
LOMDedicated (1) GbE management port
Optional NICPlease refer to our Compatible Component List for more information
Front I/O
Front I/OPower/ID/Reset Buttons
Power/ID/Status LEDs
(2) USB ports
(1) VGA port (Display Priority: First; one device one time)
Storage Controller
Optional ControllerQuanta LSI® 3008 12Gb/s SAS mezzanine, RAID 0,1,10 
Quanta LSI® 3108 12Gb/s RAID mezzanine, RAID 0, 1, 5, 6,10
(Please refer to our Compatible Component List for more information)
Power Supply
Power Supply3+1 high efficiency hot-plug 1600W PSU, 80 Plus Platinum (200V-240V)
Onboard Storage
Onboard Storage
(2) SATADOM (option)
(2) 2280/2210 SATA/PCIe M.2 (option)
Fan
Fan(4) dual rotor fans (7+1 redundant)
Video
VideoIntegrated ASPEED AST2500 8MB DDR4 video memory
System Management
System ManagementRedfish v1.0.2
IPMI v2.0 Compliant, on board "KVM over IP" support
Rear I/O
Rear I/O(2) USB 3.0 ports
(1) VGA port (Display Priority: Second; one device one time)
(1) RS232 serial port
(1) GbE RJ45 management port
(1) ID LED
(1) MicroSD slot
Operating Environment
Operating EnvironmentOperating temperature*: 5°C to 30°C (41°F to 86°F)
Non-operating temperature: -40°C to 70°C (-40°F to 158°F)
Operating relative humidity: 20% to 85%RH.
Non-operating relative humidity: 10% to 95%RH
*Certain configuration support higher operating temperature or lower CPU TDP. Please contact local sales for more detail; The performance of the system may be impacted when operating over the maximum temperature limit.
TPM
TPMTPM 2.0

Weight (Max. Configuration)
Weight (Max. Configuration)61.5 (Kg)
Datasheet 1
Title Version Date Size
D52G-4U Datasheet v20180504 04/05/2018 371 KB
CATALOGUE 1
Title Version Date Size
HS Product Catalog 1 01-10-20 6371KB
Compatible Components List 1
Title Version Date Size
D52G-4U CCL 20/03/2018 359 KB
FRU Part List 1
Title Version Date Size
D52G-4U FRU Parts List
BIOS 1
Title Version Date Size
D52G-4U BIOS 3A14.Q301 05/06/2019 23.38 MB
BMC 1
Title Version Date Size
D52G-4U BMC 4.73.00 04/07/2019 62.04 MB
Driver-Linux 1
Title Version Date Size
D52G-4U Drivers Linux
Driver-Windows 1
Title Version Date Size
D52G-4U Drivers Windows

S5G | D52G-4U

Prevail over AI and HPC challenges

  • NVLink™ support up to 300GB/s GPU to GPU communication
  • 10x dual-width 300 Watt GPU
  • 16x single-width 75 Watt GPU support
  • Diversify GPU topology to Conquer Any Type of Parallel Computing Workload
  • Up to 4x100Gb/s High Bandwidth RDMA-enabled Network to Scale Out with Efficiency
  • 8x NVMe Storage to Accelerate Deep Learning

  • $USD $15,600.00

    *RRP Pricing*

    To View Channel Discounts Please Login


System Configurator


Tags: GPU, GP-GPU-QuantaGrid-D52G-4U, Quanta, QCT, AI, HPC, Deep, Learning, Infrencing