Qualcomm puts Beachhead into artificial intelligence with Foxconn Gloria AI Edge Box

When most people think of Qualcomm, the first technologies that are likely to come to mind are the company’s leading mobile platform-on-chips for smartphones as well as the company’s end-to-end 5G connectivity solutions. Whether you are considering applications such as image recognition, speech input, natural language translation or recommendation engines, modern smartphone platforms typically also require a lot of artificial intelligence (AI) that processes horsepower. After years of developing silicon and software platform solutions for mobile AI applications, this is the reason why Qualcomm has the opportunity to bring its AI accelerator technology to other intelligent edge devices and the cloud. And that’s exactly what’s happening with Qualcomm’s Cloud AI 100 inference accelerator portfolio, as evidenced by the company’s recent joint announcement with Foxconn, one of the largest manufacturers of electronics contracts and ODMs in the world.

Commercial implementation of Qualcomm Cloud AI 100

Foxconn’s Industrial Internet division has launched a new AI-enabled machine vision platform called Gloria. Powered by Qualcomm’s Cloud AI 100 solution, the system is designed to support up to 24 high-definition cameras for video analytics applications such as traffic analytics, security monitoring and smart retail spaces. In addition, the Vietnamese AI safety power plant BKAV has announced the first customer rollout of Qualcomm Cloud AI 100-powered Foxconn Gloria in its AI portfolio of intelligent security cameras.

Qualcomm has been working with BKAV since the end of last year, and to make sense of what the company’s camera systems can do, here’s a look at an earlier product based on Qualcomm’s older Vision Intelligence Platform …

As you can see, image recognition and video analysis are effective tools for a myriad of applications, from smart cities to security, to retail and healthcare applications and more.

Under the hood of Qualcomm AI 100, Performance Claims and Foxconn Gloria

Invocation of similar silicon engine technology taken from Qualcomm’s Snapdragon Mobile Platform Hexagon and Tensor Accelerator architectures, the AI ​​100 itself uses up to 16 AI cores (AICs) and scales all the way from 15 watts up to 75 watts power envelopes. Qualcomm also notes that AIC chipsets aboard AI 100 accelerators are connected via a proprietary high-speed connection. Physically, the product is offered in both M.2 gumstick-style plug-in cards used in AI edge applications and PCI Express Gen3 / 4 accelerator cards for cloud data center servers. The AI ​​100 also has up to 144 MB of fast on-die SRAM, depending on the number of cores, and can interface with up to 32 GB of on-card or in-system LPDDR4 DRAM.

In terms of performance, Qualcomm makes some bold claims in relation to competitive solutions for applications such as image recognition (ResNet-50 model performance), for example, as you can see in the graph above, where an AI 100 20 Watt configuration provides significantly better lead throughput than even a 70 watt NVIDIA T4 GPU.

Foxconn’s Gloria AI Edge Box is a fully customized turnkey commercial device powered by Qualcomm Cloud AI 100, which BKAV will implement with its AI View camera systems, but again the application options are wide and varied. Supports both Sub-6 and mmWave (Millimeter Wave) 5G connectivity, “The local AI Edge box from Gloria manufactured by Foxconn Industrial Internet allows us to demonstrate incredible low-power performance at a competitive price point,” notes Nguyen Tu Quang, Chairman and CEO, BKAV. “Gloria will help us accelerate AI applications in companies and governments and revolutionize our work with smart cities, smart construction and agriculture, among other verticals.”

Of course, all applications, from edge devices like Foxconn Gloria to data center AI-AI processing requirements serviced by Qualcomm’s PCI Express accelerator cards, deep software tools and library support, which is a strength for Qualcomm, again require due to part of the company’s many years of AI-enabled development, implementation and support in mobile applications. Support for native AI frameworks such as TensorFlow, PyTorch and Caffe allows AI-focused developers to code on their chosen platform and then target optimization on Qualcomm Cloud AI 100.

It’s clear that the AI ​​space is hot right now, and performance-pr. Watt-optimized solutions like Foxconn’s Gloria are well-suited to the ever-growing ecosystem of smart edge devices, so the possibilities are strong for both companies. Engineering samples of Foxconn’s Gloria AI Edge Box will be available later this year, and commercial availability is expected in the second quarter of 2022. Qualcomm’s AI 100 devkit (seen above), also powered by a Snapdragon 865 platform processor with 5G connectivity supported by Qualcomms Snapdragon X55 Modem RF system, now sending to customers for development efforts now.

Give a Comment