To install click the Add extension button. That's it.

The source code for the WIKI 2 extension is being checked by specialists of the Mozilla Foundation, Google, and Apple. You could also do it yourself at any point in time.

How to transfigure the Wikipedia

Would you like Wikipedia to always look as professional and up-to-date? We have created a browser extension. It will enhance any encyclopedic page you visit with the magic of the WIKI 2 technology.

Try it — you can delete it anytime.

Install in 5 seconds
4,5
Kelly Slayton
Congratulations on this excellent venture… what a great idea!
Alexander Grigorievskiy
I use WIKI 2 every day and almost forgot how the original Wikipedia looks like.
Live Statistics
English Articles
Improved in 24 Hours
Added in 24 Hours
Languages
Recent
Show all languages
What we do. Every page goes through several hundred of perfecting techniques; in live mode. Quite the same Wikipedia. Just better.
Great Wikipedia has got greater.
.
Leo
Newton
Brights
Milds

Nvidia DGX

From Wikipedia, the free encyclopedia

A rack containing five DGX-1 supercomputers

Nvidia DGX is a line of Nvidia-produced servers and workstations which specialize in using GPGPU to accelerate deep learning applications.[1] The typical design of a DGX system is based upon a rackmount chassis with motherboard that carries high performance x86 server CPUs.[2] The main component of a DGX system is a set of 4 to 8 Nvidia Tesla GPU modules on an independent system board. DGX systems have large heatsinks and powerful fans to adequately cool thousands of watts of thermal output. The GPU modules are typically integrated into the system using a version of the SXM socket or by a PCIe x16 slot.

YouTube Encyclopedic

  • 1/3
    Views:
    48 994
    81 480
    29 853
  • FIRST LOOK: NVIDIA DGX Station A100 Unboxing
  • Checking out one of the world’s first NVIDIA DGX Station A100s
  • Distributed Training On NVIDIA DGX Station A100 | Deep Learning Tutorial 43 (Tensorflow & Python)

Transcription

Models

Pascal - Volta

DGX-1

DGX-1 servers feature 8 GPUs based on the Pascal or Volta daughter cards[3] with 128GB of total HBM2 memory, connected by an NVLink mesh network.[4] The DGX-1 was announced on 6 April in 2016.[5] All models are based on a dual socket configuration of Intel Xeon E5 CPUs, and are equipped with the following features.

  • 512 GB of DDR4-2133
  • Dual 10Gb networking
  • 4 x 1.92 TB SSDs
  • 3200W of combined power supply capability
  • 3U Rackmount Chassis

The product line is intended to bridge the gap between GPUs and AI accelerators in that the device has specific features specializing it for deep learning workloads.[6] The initial Pascal based DGX-1 delivered 170 teraflops of half precision processing,[7] while the Volta-based upgrade increased this to 960 teraflops.[8]

The DGX-1 was first available only with the Pascal based configuration, with the first generation SXM socket. The later revision of the DGX-1 offered support for first generation Volta cards via the SXM-2 socket. Nvidia offered upgrade kits that allowed users with a Pascal based DGX-1 to upgrade to a Volta based DGX-1.[9][10]

  • The Pascal based DGX-1 has two variants, one with a 16 core Intel Xeon E5-2698 V3, and one with a 20 core E5-2698 V4. Pricing for the variant equipped with an E5-2698 V4 is unavailable, the Pascal based DGX-1 with an E5-2698 V3 was priced at launch at $129,000[11]
  • The Volta based DGX-1 is equipped with an E5-2698 V4 and was priced at launch at $149,000.[11]

DGX Station

Designed as a turnkey deskside AI supercomputer, the DGX Station is a tower computer that can function completely independently without typical datacenter infrastructure such as cooling, redundant power, or 19 inch racks.

The DGX station was first available with the following specifications.[12]

  • Four Volta-based Tesla V100 accelerators, each with 16 GB of HBM2 memory
  • 480 TFLOPS FP16
  • Single Intel Xeon E5-2698 v4[13]
  • 256 GB DDR4
  • 4x 1.92 TB SSDs
  • Dual 10 Gb Ethernet

The DGX station is water-cooled to better manage the heat of almost 1500W of total system components, this allows it to keep a noise range under 35 dB under load.[14] This, among other features, made this system a compelling purchase for customers without the infrastructure to run rackmount DGX systems, which can be loud, output a lot of heat, and take up a large area. This was Nvidia's first venture into bringing high performance computing deskside, which has since remained a prominent marketing strategy for Nvidia.[15]

DGX-2

The successor of the Nvidia DGX-1 is the Nvidia DGX-2, which uses sixteen Volta-based V100 32GB (second generation) cards in a single unit. It was announced on 27 March in 2018.[16] The DGX-2 delivers 2 Petaflops with 512GB of shared memory for tackling massive datasets and uses NVSwitch for high-bandwidth internal communication. DGX-2 has a total of 512GB of HBM2 memory, a total of 1.5TB of DDR4. Also present are eight 100Gb/sec InfiniBand cards and 30.72 TB of SSD storage,[17] all enclosed within a massive 10U rackmount chassis and drawing up to 10 kW under maximum load.[18] The initial price for the DGX-2 was $399,000.[19]

The DGX-2 differs from other DGX models in that it contains two separate GPU daughterboards, each with eight GPUs. These boards are connected by an NVSwitch system that allows for full bandwidth communication across all GPUs in the system, without additional latency between boards.[18]

A higher performance variant of the DGX-2, the DGX-2H, was offered as well. The DGX-2H replaced the DGX-2's dual Intel Xeon Platinum 8168's with upgraded dual Intel Xeon Platinum 8174's. This upgrade does not increase core count per system, as both CPUs are 24 cores, nor does it enable any new functions of the system, but it does increase the base frequency of the CPUs from 2.7 GHz to 3.1 GHz.[20][21][22]

Ampere

DGX A100 Server

Announced and released on May 14, 2020. The DGX A100 was the 3rd generation of DGX server, including 8 Ampere-based A100 accelerators.[23] Also included is 15TB of PCIe gen 4 NVMe storage,[24] 1 TB of RAM, and eight Mellanox-powered 200GB/s HDR InfiniBand ConnectX-6 NICs. The DGX A100 is in a much smaller enclosure than its predecessor, the DGX-2, taking up only 6 Rack units.[25]

The DGX A100 also moved to a 64 core AMD EPYC 7742 CPU, the first DGX server to not be built with an Intel Xeon CPU. The initial price for the DGX A100 Server was $199,000.[23]

DGX Station A100

As the successor to the original DGX Station, the DGX Station A100, aims to fill the same niche as the DGX station in being a quiet, efficient, turnkey cluster-in-a-box solution that can be purchased, leased, or rented by smaller companies or individuals who want to utilize machine learning. It follows many of the design choices of the original DGX station, such as the tower orientation, single socket CPU mainboard, a new refrigerant-based cooling system, and a reduced number of accelerators compared to the corresponding rackmount DGX A100 of the same generation.[15] The price for the DGX Station A100 320G is $149,000 and $99,000 for the 160G model, Nvidia also offers Station rental at ~$9000 USD per month through partners in the US (rentacomputer.com) and Europe (iRent IT Systems) to help reduce the costs of implementing these systems at a small scale.[26][27]

The DGX Station A100 comes with two different configurations of the built in A100.

  • Four Ampere-based A100 accelerators, configured with 40GB (HBM) or 80GB (HBM2e) memory,
    thus giving a total of 160GB or 320GB resulting either in DGX Station A100 variants 160G or 320G.
  • 2.5 PFLOPS FP16
  • Single 64 Core AMD EPYC 7742
  • 512 GB DDR4
  • 1 x 1.92 TB NVMe OS drive
  • 1 x 7.68 TB U.2 NVMe Drive
  • Dual port 10Gb Ethernet
  • Single port 1Gb BMC port

Hopper

DGX H100 Server

Announced March 22, 2022[28] and planned for release in Q3 2022,[29] The DGX H100 is the 4th generation of DGX servers, built with 8 Hopper-based H100 accelerators, for a total of 32 PFLOPs of FP8 AI compute and 640GB of HBM3 Memory, an upgrade over the DGX A100s HBM2 memory. This upgrade also increases VRAM bandwidth to 3 TB/s.[30] The DGX H100 increases the rackmount size to 8U to accommodate the 700W TDP of each H100 SXM card. The DGX H100 also has two 1.92TB SSDs for Operating System storage, and 30.72 TB of Solid state storage for application data.

One more notable addition is the presence of two Nvidia Bluefield 3 DPUs,[31] and the upgrade to 400Gb/s InfiniBand via Mellanox ConnectX-7 NICs, double the bandwidth of the DGX A100. The DGX H100 uses new 'Cedar Fever' cards, each with four ConnectX-7 400GB/s controllers, and two cards per system. This gives the DGX H100 3.2Tb/s of fabric bandwidth across Infiniband.[32]

The DGX H100 has two Xeon Platinum 8480C Scalable CPUs (Codenamed Sapphire Rapids)[33] and 2 Terabytes of System Memory.[34]

The DGX H100 was priced at £379,000 or ~$482,000 USD at release.[35]

DGX GH200

Announced May, 2023, the DGX GH200 connects 32 NVIDIA Grace Hopper Superchips into a singular superchip, that consists totally of 256 H100 GPUs, 32 Grace Neoverse V2 72-core CPUs, 32 OSFT single-port ConnectX-7 VPI of with 400Gb/s InfiniBand and 16 dual-port BlueField-3 VPI with 200Gb/s of Mellanox [1] [2] . NVIDIA DGX™ GH200 is designed to handle terabyte-class models for massive recommender systems, generative AI, and graph analytics, offering 19.5TB of shared memory with linear scalability for giant AI models.[36]

DGX Helios

Announced May, 2023, the DGX Helios supercomputer features 4 DGX GH200 systems. Each is interconnected with NVIDIA Quantum-2 InfiniBand networking to supercharge data throughput for training large AI models. Helios includes 1,024 H100 GPUs.

Blackwell

DGX GB200

Announced March 2024, GB200 NVL72 connects 36 Grace Neoverse V2 72-core CPUs and 72 B100 GPUs in a rack-scale design. The GB200 NVL72 is a liquid-cooled, rack-scale solution that boasts a 72-GPU NVLink domain that acts as a single massive GPU [3]. NVIDIA DGX™ GB200 offers 13.5TB HBM3e of shared memory with linear scalability for giant AI models, less than its predecessor DGX GH200.

DGX SuperPod

The DGX Superpod is a high performance turnkey supercomputer solution provided by Nvidia using DGX hardware.[37] This tightly integrated system combines high performance DGX compute nodes with fast storage and high bandwidth networking to provide a unique plug & play solution to extremely high demand machine learning workloads. The Selene Supercomputer, at the Argonne National Laboratory, is one example of a DGX SuperPod based system.

Selene, built from 280 DGX A100 nodes, ranked 5th on the Top500 list for most powerful supercomputers at the time of its completion, and has continued to remain high in performance. This same integration is available to any customer with minimal effort on their behalf, and the new Hopper based SuperPod can scale to 32 DGX H100 nodes, for a total of 256 H100 GPUs and 64 x86 CPUs. This gives the complete SuperPod a whopping 20TB of HBM3 memory, 70.4 TB/s of bisection bandwidth, and up to 1 ExaFLOP of FP8 AI compute.[38] These SuperPods can then be further joined to create even larger supercomputers.

Eos supercomputer, designed, built, and operated by Nvidia,[39][40][41] was constructed of 18 H100 based SuperPods, totaling 576 DGX H100 systems, 500 Quantum-2 InfiniBand switches, and 360 NVLink Switches, that allow Eos to deliver 18 EFLOPs of FP8 compute, and 9 EFLOPs of FP16 compute, making Eos the 5th fastest AI supercomputer in the world, according to TOP500 (November 2023 edition).

As Nvidia does not produce any storage devices or systems, Nvidia SuperPods rely on partners to provide high performance storage. Current storage partners for Nvidia Superpods are Dell EMC, DDN, HPE, IBM, NetApp, Pavilion Data, and VAST Data.[42]

Accelerators

Comparison of accelerators used in DGX:[43][44][45]

Model Architecture Socket FP32
CUDA
cores		 FP64 cores
(excl. tensor)		 Mixed
INT32/FP32
cores		 INT32
cores		 Boost
clock		 Memory
clock		 Memory
bus width		 Memory
bandwidth		 VRAM Single
precision
(FP32)		 Double
precision
(FP64)		 INT8
(non-tensor)		 INT8
dense tensor		 INT32 FP16 FP16
dense tensor		 bfloat16
dense tensor		 TensorFloat-32
(TF32)
dense tensor		 FP64
dense tensor		 Interconnect
(NVLink)		 GPU L1 Cache L2 Cache TDP Die size Transistor
count		 Process
B100 Blackwell N/A N/A N/A N/A N/A N/A N/A N/A 8TB/sec 192GB HBM3e N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 1.8TB/sec GB200 N/A N/A N/A N/A 208 B N/A
H200 Hopper SXM5 N/A N/A N/A N/A N/A N/A N/A N/A 141GB HBM3e 67 TFLOPs 34 TFLOPs N/A 4000 TOPs N/A N/A N/A N/A 989 TFLOPs N/A 900 GB/sec GH200 N/A N/A 700 W N/A N/A TSMC 4 nm N4
H100 Hopper SXM5 16896 4608 16896 N/A 1780 MHz 4.8 Gbit/s HBM3 5120-bit 3072 GB/sec 80GB HBM3 60 TFLOPs 30 TFLOPs N/A 4000 TOPs N/A N/A 2000 TFLOPs 2000 TFLOPs 1000 TFLOPs 60 TFLOPs 900 GB/sec GH100 25344 KB (192 KB × 132) 51200 KB 700 W 814 mm2 80 B TSMC 4 nm N4
A100 80GB Ampere SXM4 6912 3456 6912 N/A 1410 MHz 3.2 Gbit/s HBM2e 5120-bit 1555 GB/sec 80GB HBM2e 19.5 TFLOPs 9.7 TFLOPs N/A 624 TOPs 19.5 TOPs 78 TFLOPs 312 TFLOPs 312 TFLOPs 156 TFLOPs 19.5 TFLOPs 600 GB/sec GA100 20736 KB (192 KB × 108) 40960 KB 400 W 826 mm2 54.2 B TSMC 7 nm N7
A100 40GB Ampere SXM4 6912 3456 6912 N/A 1410 MHz 2.4 Gbit/s HBM2 5120-bit 1555 GB/sec 40GB HBM2 19.5 TFLOPs 9.7 TFLOPs N/A 624 TOPs 19.5 TOPs 78 TFLOPs 312 TFLOPs 312 TFLOPs 156 TFLOPs 19.5 TFLOPs 600 GB/sec GA100 20736 KB (192 KB × 108) 40960 KB 400 W 826 mm2 54.2 B TSMC 7 nm N7
V100 32GB Volta SXM3 5120 2560 N/A 5120 1530 MHz 1.75 Gbit/s HBM2 4096-bit 900 GB/sec 32GB HBM2 15.7 TFLOPs 7.8 TFLOPs 62 TOPs N/A 15.7 TOPs 31.4 TFLOPs 125 TFLOPs N/A N/A N/A 300 GB/sec GV100 10240 KB (128 KB × 80) 6144 KB 350 W 815 mm2 21.1 B TSMC 12 nm FFN
V100 16GB Volta SXM2 5120 2560 N/A 5120 1530 MHz 1.75 Gbit/s HBM2 4096-bit 900 GB/sec 16GB HBM2 15.7 TFLOPs 7.8 TFLOPs 62 TOPs N/A 15.7 TOPs 31.4 TFLOPs 125 TFLOPs N/A N/A N/A 300 GB/sec GV100 10240 KB (128 KB × 80) 6144 KB 300 W 815 mm2 21.1 B TSMC 12 nm FFN
P100 Pascal SXM/SXM2 N/A 1792 3584 N/A 1480 MHz 1.4 Gbit/s HBM2 4096-bit 720 GB/sec 16GB HBM2 10.6 TFLOPs 5.3 TFLOPs N/A N/A N/A 21.2 TFLOPs N/A N/A N/A N/A 160 GB/sec GP100 1344 KB (24 KB × 56) 4096 KB 300 W 610 mm2 15.3 B TSMC 16 nm FinFET+

See also

References

  1. ^ "NVIDIA DGX-1: Deep Learning Server for AI Research". NVIDIA. Retrieved 24 March 2022.
  2. ^ "NVIDIA DGX Systems for Enterprise AI". NVIDIA. Retrieved 24 March 2022.
  3. ^ "nvidia dgx-1" (PDF). Retrieved 15 November 2023.
  4. ^ "inside pascal". 5 April 2016. Eight GPU hybrid cube mesh architecture with NVLink
  5. ^ "NVIDIA Unveils the DGX-1 HPC Server: 8 Teslas, 3U, Q2 2016".
  6. ^ "deep learning supercomputer". 5 April 2016.
  7. ^ "DGX-1 deep learning system" (PDF). NVIDIA DGX-1 Delivers 75X Faster Training...Note: Caffe benchmark with AlexNet, training 1.28M images with 90 epochs
  8. ^ "DGX Server". DGX Server. Nvidia. Retrieved 7 September 2017.
  9. ^ Volta architecture whitepaper nvidia.com
  10. ^ Use Guide nvidia.com
  11. ^ a b Oh, Nate. "NVIDIA Ships First Volta-based DGX Systems". www.anandtech.com. Retrieved 24 March 2022.
  12. ^ "CompecTA | NVIDIA DGX Station Deep Learning System". www.compecta.com. Retrieved 24 March 2022.
  13. ^ "Intel® Xeon® Processor E5-2698 v4 (50M Cache, 2.20 GHz) - Product Specifications". Intel. Retrieved 19 August 2023.
  14. ^ Supercomputer datasheet nvidia.com
  15. ^ a b "NVIDIA DGX Platform". NVIDIA. Retrieved 15 November 2023.
  16. ^ "Nvidia launches the DGX-2 with two petaFLOPS of power". 28 March 2018.
  17. ^ "NVIDIA DGX -2 for Complex AI Challenges". NVIDIA. Retrieved 24 March 2022.
  18. ^ a b Cutress, Ian. "NVIDIA's DGX-2: Sixteen Tesla V100s, 30 TB of NVMe, only $400K". www.anandtech.com. Retrieved 28 April 2022.
  19. ^ "The NVIDIA DGX-2 is the world's first 2-petaflop single server supercomputer". www.hardwarezone.com.sg. Retrieved 24 March 2022.
  20. ^ DGX2 User Guide nvidia.com
  21. ^ "Product Specifications". www.intel.com. Retrieved 28 April 2022.
  22. ^ "Product Specifications". www.intel.com. Retrieved 28 April 2022.
  23. ^ a b Ryan Smith (14 May 2020). "NVIDIA Ampere Unleashed: NVIDIA Announces New GPU Architecture, A100 GPU, and Accelerator". AnandTech.
  24. ^ Tom Warren; James Vincent (14 May 2020). "Nvidia's first Ampere GPU is designed for data centers and AI, not your PC". The Verge.
  25. ^ "Boston Labs welcomes the DGX A100 to our remote testing portfolio!". www.boston.co.uk. Retrieved 24 March 2022.
  26. ^ Mayank Sharma (13 April 2021). "Nvidia will let you rent its mini supercomputers". TechRadar. Retrieved 31 March 2022.
  27. ^ Jarred Walton (12 April 2021). "Nvidia Refreshes Expensive, Powerful DGX Station 320G and DGX Superpod". Tom's Hardware. Retrieved 28 April 2022.
  28. ^ Newsroom, NVIDIA. "NVIDIA Announces DGX H100 Systems – World's Most Advanced Enterprise AI Infrastructure". NVIDIA Newsroom Newsroom. Retrieved 24 March 2022.
  29. ^ Albert (24 March 2022). "NVIDIA H100: Overview, Specs, & Release Date | SeiMaxim". www.seimaxim.com. Retrieved 22 August 2022.
  30. ^ Walton, Jarred (22 March 2022). "Nvidia Reveals Hopper H100 GPU With 80 Billion Transistors". Tom's Hardware. Retrieved 24 March 2022.
  31. ^ Newsroom, NVIDIA. "NVIDIA Announces DGX H100 Systems – World's Most Advanced Enterprise AI Infrastructure". NVIDIA Newsroom Newsroom. Retrieved 19 April 2022.
  32. ^ servethehome (14 April 2022). "NVIDIA Cedar Fever 1.6Tbps Modules Used in the DGX H100". ServeTheHome. Retrieved 19 April 2022.
  33. ^ "NVIDIA DGX H100 Datasheet". www.nvidia.com. Retrieved 2 August 2023.
  34. ^ "NVIDIA DGX H100". NVIDIA. Retrieved 24 March 2022.
  35. ^ Every NVIDIA DGX benchmarked & power efficiency & value compared, including the latest DGX H100., retrieved 1 March 2023
  36. ^ "NVIDIA DGX GH200". NVIDIA. Retrieved 24 March 2022.
  37. ^ "NVIDIA SuperPOD Datasheet". NVIDIA. Retrieved 15 November 2023.
  38. ^ Jarred Walton (22 March 2022). "Nvidia Reveals Hopper H100 GPU With 80 Billion Transistors". Tom's Hardware. Retrieved 24 March 2022.
  39. ^ Vincent, James (22 March 2022). "Nvidia reveals H100 GPU for AI and teases 'world's fastest AI supercomputer'". The Verge. Retrieved 16 May 2022.
  40. ^ Mellor, Chris (31 March 2022). "Nvidia Eos AI supercomputer will need a monster storage system". Blocks and Files. Retrieved 21 May 2022.
  41. ^ Comment, Sebastian Moss. "Nvidia announces Eos, "world's fastest AI supercomputer"". Data Center Dynamics. Retrieved 21 May 2022.
  42. ^ Mellor, Chris (31 March 2022). "Nvidia Eos AI supercomputer will need a monster storage system". Blocks and Files. Retrieved 29 April 2022.
  43. ^ Smith, Ryan (22 March 2022). "NVIDIA Hopper GPU Architecture and H100 Accelerator Announced: Working Smarter and Harder". AnandTech.
  44. ^ Smith, Ryan (14 May 2020). "NVIDIA Ampere Unleashed: NVIDIA Announces New GPU Architecture, A100 GPU, and Accelerator". AnandTech.
  45. ^ "NVIDIA Tesla V100 tested: near unbelievable GPU power". TweakTown. 17 September 2017.
This page was last edited on 28 March 2024, at 03:30
Basis of this page is in Wikipedia. Text is available under the CC BY-SA 3.0 Unported License. Non-text media are available under their specified licenses. Wikipedia® is a registered trademark of the Wikimedia Foundation, Inc. WIKI 2 is an independent company and has no affiliation with Wikimedia Foundation.
Contact WIKI 2
Introduction
Terms of Service
Privacy Policy