A working group of AI experts in India has proposed a bold initiative to advance AI innovation in the country. The plan entails the creation of a computing infrastructure that includes 24,500 graphics processing units (GPUs) distributed across 17 centers. The primary focus is to empower startups and academia to explore and develop emerging AI technologies.
The proposal recommends allocating approximately 14,500 GPUs for AI model training and high-performance storage, with the remaining 10,000 GPUs dedicated to AI inferencing. These GPU centers will consist of clusters of computers equipped with GPUs, facilitating the training of neural networks for tasks like image and video processing.
Union Minister Rajeev Chandrasekhar has expressed interest in pursuing a public-private partnership model for manufacturing AI chips and establishing GPU infrastructure in India based on the working group’s recommendations.
The envisioned infrastructure follows a three-layered system, encompassing high-end compute infrastructure, an inference arm, and edge compute resources. These components will be strategically distributed to efficiently address diverse computational requirements, allowing users to seamlessly transition between them while harnessing their unique capabilities.
To enable smooth data exchange and collaboration, a secure distributed data grid will serve as a robust and secure framework for uploading, downloading, and exchanging large datasets and AI models.
The report acknowledges the global landscape of AI compute infrastructure, highlighting the prominent roles of the United States and China in AI development. The U.S. boasts numerous supercomputers and high-performance computing facilities operated by government institutions, academic research centers, and private entities. China has also made significant strides in AI compute infrastructure, with several supercomputers and cutting-edge facilities.
Leading companies, including NVIDIA, Intel, AMD, IBM, and Google, dominate the market share of compute infrastructure and AI. These industry giants heavily invest in research and development to create high-performance processors, GPUs, Application-Specific Integrated Circuit (ASIC) AI chips, and software frameworks.
In addition to on-premises installations, this hardware is accessible through cloud service providers like Amazon Web Services, Microsoft Azure, and Google Cloud Platform. These providers offer on-demand access to scalable AI compute resources, reducing the need for extensive infrastructure investments and promoting the widespread adoption of AI technologies. These advancements are anticipated to drive a more than fivefold increase in the global AI infrastructure market. This ambitious initiative reflects India’s commitment to fostering AI innovation and technological development within its borders.
Read Now:Earth’s Inner Core May Be ‘Softer Than Butter,’ Revealing New Mysteries
