AAEON Partners with AI Chipmaker Hailo to Enable Next-Gen AI Applications at the Edge
Edge computing requires increasingly intensive workloads for computer vision and other artificial intelligence tasks, making it increasingly important to move Deep Learning workloads from the cloud to the edge. Running AI applications at the edge ensures real-time inferencing, data privacy, and low latency for smart city, smart retail, Industry 4.0, and many other applications across various markets.
UP Bridge the Gap edge platforms pair the latest Intel processors with the Hailo-8 AI Module to increase on-device inference and execution speed of AI models. Reallocating the computing resource of AI-based tasks from the CPU/GPU to the Hailo-8 neural processing unit (NPU) provides greater flexibility and possibilities for edge computing applications.
The Hailo-8 M.2 AI Module delivers 26 tera-operations per second (TOPS) at a record-breaking average rate of 2.5 watt. This means developers can run sophisticated Deep Learning and computer vision applications at the edge while keeping power consumption at an absolute minimum.
“This partnership with Hailo benefits our clients even further with their growing demand for AI applications that run quickly and efficiently at the edge,” said Owen Wei, Business Development Manager at AAEON. “AAEON is committed to providing customers exceptional computing performance for their everyday business needs. Hailo helps us bolster our offering while enabling a future powered at the edge.”
“Enterprises require solutions that are more powerful, versatile, responsive, and secure. AI at the edge is key to meeting these demands,” said Liran Bar, VP Business Development at Hailo. “We’re proud to work with AAEON to enable a new era of edge computing, while bringing unmatched edge processing solutions to customers across industries.”
The latest UP Bridge the Gap platforms are compatible with the Hailo-8 M.2 2280 B+M key AI Module. These compact edge computing devices have impressive industrial features ranging from power-efficiency to high-performance.
Starting with the UP Squared Pro, which is based on the Intel Atom E3900 series (formerly Apollo Lake), developers enjoy power-efficient performance on a cost-effective device. The processors can reach speeds up to 2.5 GHz. UP Squared Pro supports up to 8 GB of LPDDR4 memory and offers up to 64 GB eMMC storage.
With the same compact form factor as the UP Squared Pro, the recently released UP Squared 6000 is based on the latest Intel Atom x6000E series (formerly Elkhart Lake). UP Squared 6000 is also an impressive power-efficient platform, but it offers up to 1.7 times better compute performance and twice the performance for 3D graphics as UP Squared Pro. UP Squared 6000 has a range of processor options, so it can reach speeds from 1.90 GHz to 3 GHz. UP Squared 6000 is available with up to 8 GB of LPDDR4 memory and up to 64 GB eMMC storage.
Developers looking for exceptional computing performance should opt for the UP Xtreme i11. Featuring the 11th Generation Intel Core Processors (formerly Tiger Lake), the UP Xtreme i11 can clock up to 4.4 GHz with only 28 watts TDP and 15 watts cTDP. It supports up to 64 GB with two slots for DDR4 SO-DIMM memory and offers a slot for a SATA III disk or NVMe disk.
The Hailo-8 M.2 AI Acceleration Module is now available for purchase as an add-on with the UP Squared Pro, UP Squared 6000, and UP Xtreme i11.