• November 19, 2024

Arm Tech Symposia Annual Technical Conference: The Three Pillars of AI

On November 19, 2024, the Arm Tech Symposia Annual Technical Conference opened in Shanghai. As an important annual event of Arm, the conference brought together nearly 2,000 professionals, engineers and developers from all walks of life with the theme of "Working Together to Create the Future". The meeting focused on cutting-edge fields such as Generative Artificial Intelligence (AI), Edge Computing AI, Large Language Model (LLM), Chiplet Technology (Chiplet), AI infrastructure construction and intelligent driving. It is committed to promoting the in-depth exchange and cooperation of these technologies within the Arm ecosystem and jointly promoting the development and application of AI technology.

 

In this activity, Arm deeply analyzed AI's requirements for computing power and showed how, as a computing platform provider, can help partners seize the development opportunities in the AI field through three core advantages - comprehensive computing subsystems, cloud-to-end software development support, and a closely collaborative ecosystem. Facing the future, Arm adopts a more systematic overall design idea, focusing on the collaborative optimization of hardware and software, and launches corresponding computing subsystems (CSS) according to different application scenarios to meet diversified AI computing needs.

 

At the software level, Arm emphasizes that software occupies an indispensable position in its computing platform and promises to continue to invest in software technology innovation, aiming to provide more than 20 million Arm architecture-based developers around the world with an efficient, convenient and seamless development environment covering cloud-to-edge devices, and making every effort to promote innovation and development in the software field. At the same time, in building an ecosystem, Arm, as one of the world's leading technology ecosystems, integrates ecological resources through the implementation of projects such as Arm comprehensive design, helps partners shorten product launch time, and jointly promotes industry progress.

 

In his keynote speech at the conference, Dermot O'Driscoll, vice president of product solutions in Arm Infrastructure Division, stressed that since Arm transformed into a computing platform company, Arm has continued to lead the computing revolution in the AI era through the deep integration of software and hardware and architectural innovation. It is predicted that by the end of 2025, there will be more than 100 billion arm devices with AI functions worldwide, which will lay a solid foundation for the widespread adoption of AI technology. Just as Arm was a key driver in the rise of smartphones, it will also become the core force driving the development of AI computing platforms.

 

Referring to achievements in the field of ecology, Dermot O'Driscoll further elaborated: "Based on a deep understanding of the needs of the industry and the rich experience accumulated by years of cooperation with partners, Arm continues to lead the new generation of ecosystems forward. In the year since the launch of the Arm Total Compute ecology program, its membership has expanded to 30. Through the Arm computing subsystem and the Arm comprehensive computing platform, we are accelerating the collaborative progress of software and hardware, so that our partners can develop broader and higher-level solutions more conveniently, thus effectively promoting the innovation of chip and artificial intelligence technology.

 

The conference brought together strategic partners from different fields. As an industry benchmark, they shared cutting-edge insights and strategic layout of AI, fully demonstrating the prosperity and diversity of the Arm ecosystem.

Ma Tao, vice president of research and development of Alibaba Cloud Intelligent Group and chairman of Longzard Community, deeply discussed Alibaba Cloud's insights in the AI field and introduced the new Anolis 23 processor specially designed for the cloud native era. Based on the concept of heterogeneous computing, this processor adopts software and hardware collaborative optimization, and uses extensive testing and co-creation of the open source community to effectively respond to the complex challenges brought about by the AI era.

 

In addition, Zheng Aiguo, General Manager of Lenovo Global SME Products and Solutions, Tang Yu, Chief Architect of Xiaopeng Automobile Embedded, and Zou Ting, Global Vice President of Arm China Business, also had in-depth exchanges and wonderfully shared AI opportunities and strategies in their respective fields. Zheng Aiguo pointed out that Lenovo regards Windows on Arm (WoA) as an important long-term strategic investment. It is not only committed to developing leading technical products, but will continue to strengthen cooperation with Arm, Microsoft and other partners to promote the sustainable growth of the ecosystem and provide users with a richer and higher-quality product experience.

 

Zou Ting pointed out: "China has shown exciting speed and achievements in the integration of AI technology and applications. From AI data centers, AI mobile phones and PCs, to AI cars and the intelligent Internet of Things, all industries are undergoing unprecedented rapid changes. China's partners have not only made remarkable progress in the research and development of basic technologies, but also shown strong vitality and potential in innovation at the application level. Arm will continue to deepen its cooperation with these outstanding partners to provide a smarter, efficient and smoother AI experience in different fields.

 

As an annual event specially designed for developers, the developer workshop launched for the first time this year has been widely praised. The workshop focuses on cutting-edge topics such as Windows on Arm native application development, LLM reasoning acceleration under Android and generated AI. It not only provides an in-depth technical communication platform for participants, but also greatly stimulates the innovative thinking of developers and brings new inspiration and impetus to future application development.

 

So far, the global shipment of chips based on Arm architecture has exceeded the 300 billion mark. These chips not only support the powerful computing power used to train large-scale language models in large data centers, but also drive terminal AI devices used around the world. As a result, the Arm platform has become an important cornerstone for the development of AI in the future.