Intel breaks out its Arc GPU to boost edge AI
Intel is launching a series of Intel Core Ultra, Core and Atom processors and discrete Intel Arc graphics processing units (GPUs) optimised for edge processing.
The PS series of Intel Core Ultra multichip edge processors launched at the Embedded World exhibition in Germany today boost the image classification inference performance by 5x over the 14th Gen Intel Core desktop processors by combining a discrete Arc GPU and a neural processing unit (NPU) in a package that fits into an embedded LGA socket. Intel says it is building on an installed base of more than 90,000 edge deployments with this new range of edge processors, although with an embedded focus these are not the 15th generation devices codenamed Arrow Lake or the 16th generation Lunar Lake processors.
The SoC is designed to enable generative AI (GenAI) and demanding graphics workloads at the edge for retail, education, smart cities and industrial customers, including GenAI-enabled kiosk and smart point-of-sale systems in bricks-and-mortar retailers, interactive whiteboards for enhanced in-classroom experiences and AI vision-enhanced industrial devices for manufacturing and roadside units.
The Intel Core processors for edge combine the GPU of the 13th Gen Intel Core soldered down mobile processors with LGA socket flexibility to prioritize system scalability and speed to deployment. This series of processors optimized for the edge offers up to 2.57x greater graphics performance compared to 13th Gen desktop processors. This comes from the graphics execution units alongside performance hybrid architecture with Thread Director and an LGA socket-based design offering customers more edge AI and graphics performance without sacrificing hardware setup flexibility.
- Modular open edge platform scales AI applications
- Embedded system module targets the AI IPC market
- Altran and Intel open up edge AI telecoms platform
The Atom x7000C Series processors increase the base frequency of up to eight Efficient-cores to 2.4GHz to boost packet processing throughput for enterprise networking and telecommunications devices. This enables telecommunications businesses to use built-in deep learning inference capabilities to support the detection of zero-day threats, boost packet and control plane processing for OpenSSL/IPSec using native instruction sets, and leverage Intel security features to harden networks.
The Atom x7000RE Series is aimed more at industrial and manufacturing end users with up to 32 graphics execution units in a ruggedized 6W-12W BGA package offering up to 9.83x image classification performance compared with the previous x6000RE Series. These supports fanless designs to enable Industry 4.0 automation for key use cases in AI-automated tending, warehouse AMR, in-line visual inspection for quality control and ruggedized industrial PC scenarios.
Finally the Arc GPU for Edge boosts performance and edge AI capabilities on legacy Intel Core systems as a discrete GPU providing accelerated AI and media and graphics processing power. Perhaps ironically, Intel also aims to eliminate vendor lock-in with an open, standards-based software stack to offer choice and flexibility when building high-performance AI applications and solutions.
“This next generation of Intel edge-optimized processors and discrete GPUs unleashes powerful AI capabilities to help businesses more seamlessly incorporate AI alongside compute, media and graphics workloads. From manufacturing to healthcare, Intel’s extensive edge AI experience and breadth and depth of edge-ready silicon and software help our customers deliver AI where they need it most for better business outcomes,” said Dan Rodriguez, Intel corporate vice president and general manager of Network and Edge Solutions Group
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
