The modern graphics processing unit (GPU) started out as an accelerator for Windows video games, but over the last 20 years has morphed into an enterprise server processor for high-performance computing and artificial-intelligence applications.\nNow GPUs are at the tip of the performance spear used in supercomputing, AI training and inference, drug research, financial modeling, and medical imaging. They have also been applied to more mainstream tasks for situations when CPUs just aren\u2019t fast enough, as in GPU-powered relational databases.\n\nAs the demand for GPUs grows, so will the competition among vendors making GPUs for servers, and there are just three: Nvidia, AMD, and (soon) Intel. Intel has tried and failed twice to come up with an alternative to the others\u2019 GPUs but is taking another run at it.\nThe importance of GPUs in data centers\nThese three vendors recognize the demand for GPUs in data centers as a growing opportunity. That\u2019s because GPUs are better suited than CPUs for handling many of the calculations required by AI and machine learning in enterprise data centers and hyperscaler networks. CPUs can handle the work; it just takes them longer.\nBecause GPUs are designed to solve complex mathematical problems in parallel by breaking them into separate tasks that they work on at the same time, they solving them more quickly. To accomplish this, they have multiple cores, many more than the general-purpose CPU. For example, Intel\u2019s Xeon server CPUs have up to 28 cores, while AMD\u2019s Epyc server CPUs have up to 64. By contrast Nvidia\u2019s current GPU generation, Ampere, has 6,912 cores, all operating in parallel to do one thing: math processing, specifically floating-point math.\nPerformance of GPUs is measured in how many of these floating-point math operations they can perform per second or FLOPS. This number sometimes specifies the standardized floating-point format in use when the measure is made, such as FP64.\nSo what does the year hold for server GPUs? Quite a bit as it turns out. Nvidia, AMD, and Intel have laid their cards on the table about their immediate plans, and it looks like this will be a stiff competition. Here\u2019s a look at what Nvidia, AMD, and Intel have in store.\nNvidia\nNvidia laid out its GPU roadmap for the year in March with the announcement of its Hopper GPU architecture, claiming that, depending on use, it can deliver three to six times the performance of its previous architecture, Ampere, which weighs in at 9.7 TFLOPS of FP64. Nvidia says the Hopper H100 will top out at 60TFLOPS of FP64 performance.\nLike previous GPUs, the Hopper H100 GPU can operate as a standalone processor running on an add-in PCI Express board in a server. But Nvidia will also pair it with a CPU on a custom Arm processor called Grace that it developed and expects have available in 2023.\nFor Hopper, Nvidia did more than just amp up the GPU processor. It also modified low-power double data rate (LPDDR) 5 memory\u2014normally used in smart phones\u2014to create LPDDR5X. It supports error-correction code (ECC) and twice the memory bandwidth of traditional DDR5 memory, for 1TBps of throughput.\nAlong with Hopper, Nvidia announced NVLink 4, its latest GPU-to-GPU interconnect. NVLink 4C2C allows Hopper GPUs to talk to each other directly with a maximum total bandwidth of 900GBs\u2014seven times faster than if they connected through a PCIe Gen5 bus.\n\u201cIf you think about data-center products, you have three components, and they have to all move forward at the same pace. That\u2019s memory, the processor, and communications,\u201d said Jon Peddie, president of Jon Peddie Research. \u201cAnd Nvidia has done that with Hopper. Those three technologies don\u2019t move in synchronization, but Nvidia has managed to do it.\u201d\nNvidia plans to ship the Hopper GPU starting in the third quarter of 2022. OEM partners include Atos, BOXX Technologies, Cisco, Dell Technologies, Fujitsu, GIGABYTE, H3C, Hewlett Packard Enterprise, Inspur, Lenovo, Nettrix, and Supermicro.\nDue to ongoing supply pressures at its chipmaker TSMC, Nvidia opened the door to possibly working with Intel\u2019s foundry business, but cautioned that such a deal would be years away.\nAMD\nAMD has the wind at its back. Sales are increasing quarter to quarter, its x86 CPU market share is growing, and in February it completed its acquisition of Xilinx and its field-programmable gate arrays (FPGA), adaptive systems on a chip (SoC), AI engines, and software expertise. It\u2019s expected that AMD will launch its Zen 4 CPU by the end of 2022.\nAMD\u2019s new gaming GPUs based on its RDNA 3 architecture are also due out this year. AMD has been tight lipped about RDNA 3 specs, but gaming-enthusiast bloggers have circulated unconfirmed rumors of a 50% to 60% performance gain over RDNA 2.\nIn the meantime, AMD has begun shipping the Instinct MI250 line of GPU accelerators for enterprise computing, considerably faster than the previous MI100 series. The memory bus has doubled from 4096 bits to 8192 bits, memory bandwidth has more than doubled to 3.2TBps from 1.23TBps, and performance has more than quadrupled from 11.5 TFLOPS of FP64 performance to 47.9TFLOPS. That\u2019s slower than AMD\u2019s Hopper 60TFLOPS, but it\u2019s still competitive.\nDaniel Newman, principal analyst with Futurum Research, said AMD\u2019s opportunity to grab market share will come as the AI market grows. And he said he believes that AMD\u2019s success with the CPU market could help its GPU sales. \u201cWhat AMD has really created over the past five, seven years is a pretty strong loyalty that can possibly carry over,\u201d he said. \u201cThe question is, can they grow AI\/HPC market share significantly?\u201d\nHe said the answer could be, \u201cYes,\u201d because the company has been extremely good at finding market opportunities and managing its supply chain in order to deliver on its goals. And with CEO Lisa Su at the helm, \u201cI find it very difficult to rule out AMD in any area in which they decided to compete at this point,\u201d he said.\nJonathan Cassell, principal analyst for advanced computing, AI, and IoT at Omdia, said he feels AMD\u2019s success with its Epyc server CPUs will provide an opening for the Instinct processor.\n\u201cI think that over time, we can see AMD leverage its success over on the data-center microprocessor side and use that is an in to get companies to take a look at [Instinct]. I think we\u2019ll be seeing AMD trying to leverage its relationships with customers to try to expand its presence out there,\u201d he said.\nInstinct has been shipping since Q1 2022. So far, its highest profile use case has been with a supercomputer at Oak Ridge National Labs, which packed a lot of performance into a very small space. But the labs are also building an all-AMD exascale supercomputer called Frontier, due for deployment later this year. OEM partners shipping products with Instinct include ASUS, ATOS, Dell Technologies, Gigabyte, Hewlett Packard Enterprise (HPE), Lenovo, Penguin Computing, and Supermicro.\nIntel\nIntel has long struggled to make anything but basic integrated GPUs for its desktop CPUs. For desktops it has its new Intel Xe line while the server equivalent is known as the Intel Server GPU.\nNow the company says it will enter the data-center GPU field this year with a processor code-named Ponte Vecchio that reportedly delivers 45TFLOPS at FP64\u2014almost the same as AMD\u2019s MI250 and 25% behind Nvidia\u2019s Hopper.\n\u201cIt\u2019s really going to disrupt the environment,\u201d said Peddie. \u201cFrom what they have told us\u2014and we\u2019ve heard from rumors and other leaks\u2014it\u2019s very scalable.\u201d Ponte Vecchio is due out later this year.\nNewman has also heard positive things about Ponte Vecchio, but said the real opportunity for Intel is with its oneAPI software strategy.\noneAPI is a unifying software-development platform the company is working on that is designed to pick the most appropriate type of silicon Intel makes\u2014x86, GPU, FPGA, AI processors\u2014when compiling applications rather than forcing the developer to pick one type of silicon and code to it. It also provides a number of API libraries for functions like video processing, communications, analytics, and neural networks.\nThis abstraction eliminates the need to determine the best processor to target, as well as the need to work with different tools, libraries, and programming languages. So rather than coding to a specific processor in a specific language, developers can focus on the business logic and write in Data Parallel C++ (DPC++), an open-source variant of C++ designed specifically for data parallelism and heterogeneous programming.\nOne factor that separates Intel from Nvidia and AMD is where it makes its chips. While the others use Taiwan chip maker TSMC, Intel manufactures many of its own chips in the US, with other factories in Ireland, Malaysia, and Israel. And in has big plans to build more in the US. That gives it certain advantages, Cassell said. \u201cThe control [it has] of its own manufacturing gives it a control of its destiny, in a certain way,\u201d he said. "I see these things as assets for the company.\u201d\nIn the end, said Newman, the competition among Nvidia, AMD, and Intel could come down to a software race. \u201cIf you asked [Nvidia\u2019s] top engineers, they\u2019ll say we\u2019re not a chip company. We\u2019re a software company. I really do believe that Intel has not really thought like a software company about AI up to now, but if they can get [oneAPI] right, I see some real opportunity there,\u201d he said.