NVIDIA acquires ARM: Creating a next-generation AI platform

NVIDIA announced Sept. 14 an agreement to acquire ARM holdings from SoftBank for $40 billion, subject to regulatory approval in the U.S., the U.K., the European Union and China. The acquisition has been rumored for several weeks, but the announcement generated negative comments from ARM customers. The two companies’ IP portfolios complement each other, especially in the context of rapidly growing AI workloads. TBR believes the combined company can successfully create new integrated AI hardware platforms, while growing profitable in each former company’s primary business, graphics processors for NVIDIA and mobile CPUs for ARM.

Complementary IP and different business models

ARM is in the CPU business. NVIDIA is in the graphics processing unit (GPU) business, and NVIDIA GPUs are increasingly used in non-graphics AI processing applications. Both companies rely on microprocessor design to deliver value and grow their businesses, but the way each company monetizes its IP is very different. NVIDIA is a traditional product-based business; it makes processors and boards that it sells to equipment manufacturers and to cloud service providers. ARM follows a licensing model; it sells the rights to use its designs and instruction sets to equipment manufacturers that often modify the ARM designs to meet their needs.

One concern of current ARM customers is that NVIDIA will eventually move ARM to a product model; only NVIDIA will make hardware that incorporates ARM designs, shutting off customers’ ability to customize ARM-based chips. This would be a disaster for the major mobile OEMS, including industry behemoths Apple and Samsung. ARM chips power virtually all smartphones and tablets, and mobile vendors rely on derivative ARM designs for differentiated products. Apple makes its own modifications and recently announced that its PCs will be migrated from Intel to ARM processors, allowing the company to have a uniform hardware platform for all its major products. Samsung designs its own ARM processors but relies on third-party ARM designer Qualcomm for many of its products. To make matters more confusing, Samsung manufactures both Qualcomm and Apple processors.

NVIDIA announced that it would continue the current ARM licensing business model and, in fact, would license some of its GPU IP in the same manner. Nevertheless, ARM customers are concerned because strategically vital licensed IP would now be owned by a hardware vendor. TBR believes the ARM licensing model will continue for ARM designs and the same model will greatly benefit NVIDIA’s GPU business as well.

NVIDIA is transitioning from graphics to AI

NVIDIA is the dominant vendor in GPUs, and for that reason, if its processors were used only for graphics, its growth would be limited to the growth of graphics applications. GPUs, however, are also well-suited for AI deep learning applications because both graphics and deep learning rely on massively parallel processing.

2Q20 is a crossover quarter. For the first time, NVIDIA data center revenue, which is almost all AI, was greater than revenue from graphics applications in PCs. NVIDIA data center revenue grew 167% year-to-year; NVIDIA will soon be dominated by AI applications in data centers. There is competition in AI processors from Google’s tensor processing unit (TPU) and from field-programmable gate arrays (FPGAs), as well as several new AI processing entrants, including two from Intel. Nevertheless, NVIDIA enjoys an enormous lead in a very rapidly growing business.

GPUs and CPUs working together

GPUs and CPUs coexist. Every device that uses GPUs for AI needs CPUs for all the other required processing. In data centers, the CPU is now almost always an Intel product. While ARM designs are increasingly powerful, as illustrated by Apple’s decision to use them for PCs, they are not yet used widely for data center devices. Where the GPU is doing most of the work, however, ARM-NVIDIA designs could be quite viable. ARM-NVIDIA designs would also work well in edge devices. This synergy positions NVIDIA well in a world where deep learning is becoming increasingly important.

Applications for deep learning are becoming more diverse, creating a variety of settings and requirements for CPU-GPU platforms. This proliferation of design requirements is a challenge for a product-based company like NVIDIA. The ARM licensing business model fits this diversifying market very well. TBR believes NVIDIA will first experiment with the licensing of older GPU designs, but then move rapidly to licensing GPU IP for all AI applications, greatly accelerating adoption of NVIDIA designs for AI and inhibiting growth of competing AI chip designs.

The ARM acquisition will accelerate AI

While NVIDIA and ARM are not competitors, therefore reducing anti-trust concerns, many parties have expressed concerns about this acquisition. Both companies are very important, with NVIDIA dominating AI processors and ARM monopolizing mobile CPUs. There are also concerns about a U.S. company controlling these two critical components. In the U.K., there is concern about the loss of jobs. TBR, however, believes this union will prove beneficial, certainly to the combined company, but also to other companies basing their business on the growth of AI.

AI chips: Explosive growth of deep learning is leading to rapid evolution of diverse, dedicated processors

Artificial intelligence (AI) utilization has been accelerating rapidly for more than 10 years, as decreases in memory, storage and computation cost have made an increasing number of applications cost-effective. The technique of deep learning has emerged as the most useful. Large public websites such as Facebook (Nasdaq: FB) and Amazon (Nasdaq: AMZN), with enormous stores of data on user behavior and a clear benefit from influencing user behavior, were among the earliest adopters and continue to expand such techniques. Publicly visible applications include speech recognition, natural language processing and image recognition. Other high-value applications include network threat detection, credit fraud detection and pharmaceutical research.

Deep learning techniques are based on neural networks, inspired by animal brain structure. Neural networks perform successive computations on large amounts of data. Each iteration operates on the results of the prior computation, which is why the process is called “deep.” Deep learning relies on large amounts computation. In fact, deep learning techniques are well known; the recent growth is driven by decreasing costs of data acquisition, data transmission, data storage and computation. The new processors all aim to lower the cost of computation.

The new chips are less costly than CPUs for running deep learning workloads

Each computation is limited and tends to require relatively low precision, necessitating fewer bits than found in typical CPU operations. Deep learning computations are mostly tensor operations — predominantly matrix multiplication — and parallel tensor processing is the heart of many specialized AI chips. Traditional CPUs are relatively inefficient in carrying out this kind of processing. They cannot process many operations at the same time, and they deliver precision and capacity for complex computations that are not needed.

Nvidia (Nasdaq: NVDA) GPUs led the wave of new processors. In 2012, Google announced that its Google Brain deep learning project to recognize images of cats was powered by Nvidia GPUs, resulting in a hundredfold improvement in performance over conventional CPUs. With this kind of endorsement and with the widespread acceptance of the importance of deep learning, many companies, large and small, are following the money and investing in new types of processors. It is not certain that the GPU will be a long-term winner; successful applications of FPGAs and TPUs are plentiful.

Intel: Optimizing its scale advantage for Business of One flexibility

TBR perspective

Usually sound business execution of world-class engineering, coupled with world-class monolithic manufacturing, has made Intel a dominant force around which technology businesses have orbited for decades. Intel’s dominance has been baked in the PC and server form factors, while ever smaller price points and form factors have shifted end-customer purchase criteria from computational performance specifications to business outcomes and user experiences.

Intel’s success has broadly expanded IT to address business problems and reshape our personal lives. Intel’s revenue growth prospects have diminished as its innovation has continued to increase the capacity and shrink the form factors and unit cost of its products. Intel delivers mature components that are embedded in mature products. Nevertheless, Intel thrives. The company has made mistakes, though, such as failing to address the mobile market. Intel’s capital- and engineering-intensive business requires it place large bets on its vision of the future. Now, facing waves of innovation in artificial intelligence (AI), Internet of Things (IoT) and processor design, Intel is, in effect, rearchitecting the company to reduce its dependence on the CPU, and thereby expand its market.

The key to Intel’s new architecture is companywide integration. Intel has always had more products and technologies, including video, networking, storage and memory silicon, than CPUs. As silicon becomes more diversified and is embedded in an increasing number of devices, Intel aims to create, along with customers, a far larger variety of solutions, often at a much smaller scale than the company’s monolithic products. To capitalize on the company’s enormous intellectual property, Intel must break down silos within the company. This will result in products that will often benefit from breaking down silos in silicon by facilitating the integration of computation, storage and communications.

The cultural challenge Intel will face will be in orchestrating and timing the various development teams such that the innovation cycles come together in world-class packages of tightly coupled compute, storage and networking form factors to power the smallest of edge compute instances and the largest of the high-performance computing (HPC) instances. The necessary work of rearchitecting the sales and marketing organizations remains for the next CEO, who has not yet been named, but the task is far less daunting than coordinating development and manufacture.

The thread that will stitch together these instances in the multicloud, always-on world of compute will be software. Software made interoperable through a “pruning,” as Intel Chief Engineering Officer and Technology, Systems Architecture & Client Group President Murthy Renduchintala described it, of the existing assets and frameworks into a cogent set of frameworks and tool sets to power innovation and optimize these scaled designs for specific workloads powered by AI is fed by voice and video as much as they have been fed by human interaction through keyboards in the past.


Intel Analyst Summit: Intel (Nasdaq: INTC) hosted an analyst event for the first time in four years to outline its technology road maps through 2021 and to articulate the business and cultural changes it believes are necessary for it to capitalize on the growing business opportunity Moore’s Law economics has unleashed. The senior leadership team gave about 50 analysts very detailed and frank briefings under a nondisclosure agreement (NDA), with ample time for follow-up conversations throughout the event.