Why can NVIDIA's deep learning supercomputer spike Intel?

The role of hardware and algorithms in the field of artificial intelligence can be said to be half, and at the chip level, the industry has almost the same view - GPU is much more important in artificial intelligence deep learning algorithms than CPU, which is why NVIDIA The limelight in the field of artificial intelligence has even overshadowed Intel. There is no doubt that GPU is the most popular method for training deep learning neural networks. This solution has been favored by companies such as Google, Microsoft, IBM, Toyota and Baidu. Therefore, GPU manufacturers have gradually become worshipped by enterprises in the last two years. Object.

Why can NVIDIA's deep learning supercomputer spike Intel?

As the absolute leader in the GPU field, NVIDIA has recently made frequent moves. Earlier this year, the company introduced the Tesla P100 GPU for deep neural networks and released the NVIDIA DGX-1, a single-chassis deep learning supercomputer based on the GPU.

Now that this deep learning supercomputer has been released, NVIDIA CEO Huang Renxun has delivered DGX-1 to the artificial intelligence project OpenAI founded by Musk, what will OpenAI use DGX-1? how to use? It's not known, but let's talk about what this deep learning supercomputer is. It has something to offer.

Why can NVIDIA's deep learning supercomputer spike Intel?

What is a deep learning supercomputer?

As the name implies, the deep learning supercomputer is a combination of deep learning and supercomputer. The well-known “Tianhe No.1” and “Tianhe No.2” are supercomputers. Of course, not only that, but also high performance computing (HPC). Computers can be counted as supercomputers, such as NVIDIA's Tesla series.

Because deep learning neural networks, especially hundreds of thousands of layers of neural networks, require a high level of computational and throughput capabilities, GPUs have a natural advantage in dealing with complex operations: they have excellent floating-point computing performance, while ensuring classification and Convolution performance and accuracy.

Why can NVIDIA's deep learning supercomputer spike Intel?

So supercomputers equipped with GPUs have become the only choice for training various deep neural networks. For example, in the Google Brain project, 12 GPUs are configured in 3 machines, and the performance reaches the level of a CPU cluster containing 1000 nodes.

DDR4 Desktop Memory

OEM Laptop Ram manufacturer Memoria Ram ddr 4 4g 2666 memory sodimm 4gb ddr4 2666mhz dual channel RAM for laptop 16c PC4 21300 1Rx8

• 240-pin, unbuffered dual in-line memory module (UDIMM)

• VDD = VDDQ = 1.35V (1.283–1.45V)
• Fast data transfer rates: PC3-14900, PC3-12800, or PC3-10600 • 4GB , 8GB
• Nominal and dynamic on-die termination (ODT) for data, strobe, and mask signals
• Backward-compatible to VDD = VDDQ = 1.5V ± 0.75V • VDDSPD = 3.0–3.6V • Reset pin for improved system stability • Dual-rank
• Adjustable data-output drive strength
• Fixed burst chop (BC) of 4 and burst length (BL) of 8 via the mode register set (MRS) • Serial presence-detect (SPD) EEPROM • Gold edge contacts • Halogen-free • Addresses are mirrored for second rank


Ddr4 Desktop Memory,Ddr3 4Gb 1600Mhz,Ddr3 4Gb Ram Game Memory,Ddr3 8Gb Desktop Memory

MICROBITS TECHNOLOGY LIMITED , https://www.hkmicrobits.com

Posted on