NEW TAIPEI CITY, March 23, 2023 -- Aetina, a leading provider of state-of-the-art AI solutions, is participating in NVIDIA GTC, a global conference for the era of AI and the metaverse running through March 23, showcases its latest solutions that leverage the NVIDIA® Jetson™ platform for edge AI and robotics, the NVIDIA AI Enterprise software suite, and NVIDIA GPUs. With the solutions powered by NVIDIA AI, Aetina can help developers create various types of vertical AI and enable smoother, easier AI adoption.
MXM Modules Powered by the NVIDIA RTX Family
Aetina offers MXM modules powered by NVIDIA Ampere architecture GPUs from the NVIDIA RTX™ family. These models of MXM modules include M3A4500-WP, M3A2000-VY, and M3A1000-PP, which deliver outstanding AI performance and are suitable for any AI systems that involve heavy inference workloads and real-time data analytics. They support CUDA Compute version 8.6, OpenGL 4.6, Vulkan 1.2, DirectX 12 Ultimate, Shader Model 7.0, as well as Windows 10/11 64-bit and Linux 64-bit operating systems.
For the MXM modules, Aetina provides optional conformal coating service in case of harsh system environments, and ensures the 5-year availability of the modules as well as exclusive technical support to help its clients reduce maintenance costs for their long-term AI projects.
Systems and Platforms Built with NVIDIA Jetson
As an Elite member of the NVIDIA Partner Network, Aetina offers many edge computing systems and platforms—small-sized embedded computers—that support different NVIDIA Jetson system-on-modules (SoMs), including the Jetson Orin series: Jetson AGX Orin, Jetson Orin NX, and Jetson Orin Nano. Aetina’s newly released NVIDIA Jetson-based embedded computer models—AIB-SO21/31, AIB-SN31/41 ,AIE-KO21/31, AIE-KN31/41 , AIB-MO22/32, AIB-MN32/42, AIE-KO22/32, and AIE-KN32/42 —support either an Orin NX SoM or an Orin Nano SoM. These systems and platforms, part of Aetina’s DeviceEdge family, are compatible with several kinds of sensors and hardware expansions, enabling them to be easily integrated into larger AI-driven systems and act as a completely standalone AI computer at the edge.
In addition to its standard models, Aetina provides customization services for carrier boards, development kits, and chassis with on-demand I/O requirements and thermal designs that fit into special or harsh environments.
Solutions Based on NVIDIA AI Enterprise Suite and NVIDIA Fleet Command
Aetina helps system integrators bring AI to business owners by leveraging the NVIDIA AI Enterprise software suite and the NVIDIA Fleet Command™ platform. With these NVIDIA software tools, system integrators can develop AI applications faster, from AI model training and fine-tuning to deploying on edge devices. The integrators can also find pretrained AI models for specific verticals from the NVIDIA NGC catalog, which helps streamline their development processes.
Aetina’s x86 AI platforms—AIS-D422-A1 and AIP-FQ47-B1 from the Aetina SuperEdge and MegaEdge family, respectively—feature a powerful CPU and are compatible with standard graphic cards for computing power upgrades. The platforms, paired with the NVIDIA A2 Tensor Core GPU and the NVIDIA RTX A4500 graphics card, are NVIDIA-Certified Systems™, ideal for use with NVIDIA software tools to run AI training and inference with better performance.
Aetina’s NVIDIA-Certified System Tested with NVIDIA L4 Tensor Core GPU
Aetina’s SuperEdge AIS-D422-A1—an AI platform ideal for AI model training—is a listed member of NVIDIA-Certified Systems 3.0 that has been tested with the new NVIDIA L4 Tensor Core GPU at an industrial edge level. The platform also supports NVIDIA T4 Tensor Core GPU and NVIDIA A2 Tensor Core GPU as an NVIDIA-Certified System to provide the highest level of performance and reliability.
The AIS-D422-A1 and the NVIDIA L4 Tensor Core GPU together enable developers to accelerate AI inference for different applications on NVIDIA Triton™ Inference Server, improving the AI analytic accuracy and efficiency to make AI adoption more valuable for business owners.