This infrastructure is supported by multi-layered connectivity, ranging from low-power networks to 5G, and an advanced data platform. At the core of its capabilities is a supercomputer with 192 cores and 12TB of memory, enabling massive data processing. The infrastructure also features a parallel computing cluster with 960 cores and an NVIDIA DGX2 server, offering 2 petaFLOPS of performance and 512 GB of GPU memory, ideal for handling complex AI models and large datasets.
The Animal Production Laboratory focuses on the use of NIRS for nondestructive testing of feed, meats, oils, and other agricultural products. It is involved in national and international research projects aimed at enhancing food integrity and traceability. NIRS enables rapid, efficient analysis, critical for ensuring product quality and safety throughout the food chain.
GRADIANT's 5G lab is an experimental hub for innovating and validating new technologies. It boasts a full-fledged cellular network, including Core Network, Radio Access protocols, and Edge computing. Through end-to-end virtualisation, it leverages general-purpose hardware and open-source software libraries to deploy various network and radio functions. Equipped with computing resources, SDR platforms, and high-bandwidth signal transmission elements, it serves as a versatile 5G base station and terminal.
The Gradiant test field spans over 20 hectares and is dedicated to conducting operational tests with autonomous vehicles (UAVs, UGVs, etc.). Due to the extensive terrain available, the test field combines forest areas, fields, ponds, and even interactions with local flora and fauna. This multidisciplinary environment makes it ideal for addressing tests across multiple agricultural sectors. The area has acceptable connectivity through 4G, although Gradiant can provide its own 4G/5G station to enhance connectivity.
GRADIANT's AI lab is a cutting-edge facility for developing and validating artificial intelligence solutions. It features a comprehensive infrastructure for AI experimentation, including high-performance computing clusters, specialised GPU arrays, and advanced model serving platforms. Through end-to-end MLOps integration, it leverages scalable computing resources and open-source frameworks for deploying various AI models and automation pipelines.
