The mobile silicon giant has recently released a few details and shipping plans of the high-performance AI inference accelerator – Cloud AI 100. It was launched almost 18 months ago in April 2019 and is in the production stage currently.
Qualcomm is known worldwide for designing and manufacturing wireless telecommunications products, dominates the smartphone processor market with its Snapdragon line. It further sees Cloud AI 100, making a way into the edge server market. This new addition in Qualcomm’s family is set to deploy in the first half of 2021.
Qualcomm has released a few raw details of the Cloud AI 100 chip that will be available in three types of cards. They are a dual M.2 edge (DM.2e) form factor that offers more than 50 TOPS at 15 Watts, a dual M.2 (DM.2) card configured for 200 TOPS at 25 Watts, and a PCIe card which comes in around 400 TOPS at 75 Watts.
Qualcomm is the driving force behind 5G expansion and seems to get into the AI accelerator market. The graph displayed in Qualcomm’s media briefing portrays the PCI card version of Cloud AI 100 outperforming other solutions with the only fraction of power.
“At Qualcomm, we have a long heritage of AI R&D,” said John Kehrli, senior director of product management for Qualcomm. “We’re actually in our fifth-generation solution from the mobile side, we have over 11 years of very active R&D. So we are leveraging that knowledge, that industry expertise, but this is a different AI core, it’s not the same as mobile, but we are leveraging from that space.”
Being in the leader in mobile technology, Qualcomm has strategically engineered this AI solution. The Qualcomm Cloud AI 100 is designed to support AI solutions for multiple environments such as datacenter, cloud edge, edge appliance, and 5G infrastructure. It will also accelerate the AI experience and inferencing – including low power consumption, scale, process node leadership, and signal processing expertise, for data center providers. The advanced signal processing offers a complete system solution for AI processing up to 24 simultaneous 1080p video streams along with 5G connectivity.
“Qualcomm Technologies is well-positioned to support complete edge-to-cloud high performance AI solutions that lead the industry in performance per watt,” said Keith Kressin, senior vice president and general manager, computing and edge cloud, Qualcomm Technologies. “Qualcomm Cloud AI 100 is now shipping to select worldwide customers, and we look forward to seeing commercial products launch in the first half of 2021.”
The amalgamation of 5G and AI has stimulated the edge computing innovation. “This market is continuing to grow faster than we thought 18 months ago. The 5G edge box is a greenfield opportunity for us.” reported John Kehrli, senior director product management for Qualcomm’s Cloud AI 100.
“I expect our first commercial deployments to be more on the edge side than on the data center side, where there’s a much, much longer cycle to get it into production,” he said. “That’s not to say we don’t have significant traction and opportunities there, but I expect more that 5G edge deployments will be much faster and things like our pre-certified 5G module makes that much easier. A lot of these customers that we work with are not your traditional mobile customers. So providing them a solution that’s already pre-certified, they can quickly go to market. So [applications] in that space will come up faster.” said Kehrli.
Qualcomm is placing its Cloud AI 100 for AI inference in four key markets. They are data centers outside the cloud, ADAS, 5G edge boxes, and 5G infrastructure.
Qualcomm Cloud AI 100 software aims to connect the world through technology, to lay the foundation of a world where everything and everyone can communicate and interact seamlessly. It provides a comprehensive stack of tools and supports popular frameworks, including Tensorflow, PyTorch, Caffe, and runtimes such as GLOW and ONNX. The software suite includes the Qualcomm Cloud AI 100 Applications SDK with compiler and simulator and the Platform SDK with runtimes, APIs, kernel drivers, and tools.
Moving ahead, Qualcomm intents to facilitate running inference on the edge cloud faster and efficiently and address the specific requirement in the cloud environment.