Elma, a tech company based in Fremont, California, has just released a new rugged embedded computing system for machine learning and artificial intelligence (AI). The JetSys-5330 is specifically designed for use in harsh conditions, such as the heat and dust of construction jobsites or the remote locations of mining operations.
The computer boasts a powerful NVIDIA Jetson AGX Orin system-on-module, which delivers over 275 tera operations per second worth of computing power. Additionally, expansion slots allow for custom I/O and storage when necessary.
This new release is sure to be a hit in a range of industries, including robotics, autonomous vehicles, and agriculture. It is also perfect for wireless multi-access edge computing. Essentially, the JetSys-5330 is suited to any application where AI and deep learning are required in rugged environments, making it a versatile and much-needed piece of technology.
The NVIDIA Jetson AGX Orin module includes an impressive 2048-core NVIDIA Ampere architecture graphics processing unit (GPU), 64 Tensor Cores, and a 12-core Arm Cortex-A78AE. It also has video encoder/decoder, deep learning, and vision accelerators, not to mention an impressive 204-gigabyte-per-second memory bandwidth.
To power the NVIDIA Jetson AGX Orin module, Elma’s JetKit-3110 carrier card has been added to the mix. This card provides power and interfaces for the module, as well as three mini-PCI Express slots and two M.2 slots for custom I/O or storage expansion.
The computer also has impressive rugged features, including an IP67 rating and qualifications like MIL-STD-810G, MIL-HDBK-704F, and MIL-STD-1275D, making it a reliable option in high-pressure environments. Depending upon the configuration, the power consumption ranges from 53 to 130 Watts, with maximum weight being 16 pounds.
Overall, the JetSys-5330 is a much-needed addition to the AI and deep learning world, offering cutting-edge technology in even the harshest and most challenging conditions.
