Iceotope, a global leader in precision liquid cooling (PLC), has announced the launch of KUL AI, a new solution to deliver the promise of AI everywhere and offers significant operational advantages where enhanced thermal management and maximum server performance are critical.
KUL AI features an 8-GPU Gigabyte G293 data centre server-based solution integrated with Iceotope’s precision liquid cooling and powered by Intel Xeon Scalable processors – the most powerful server integrated by Iceotope to date.
Designed to support dense GPU compute, the 8-GPU G293 carries NVIDIA Certified-Solutions accreditation and is optimised by design for liquid cooling with dielectric fluids. KUL AI ensures uninterrupted, reliable compute performance by maintaining optimal temperatures, protecting critical IT components and minimising failure rates, even during sustained GPU operations.
KUL AI enables AI everywhere
The surge in power consumption and sheer volume of data produced by new technologies including Artificial Intelligence (AI), High-Performance Computing (HPC) and Machine Learning (ML), poses significant challenges for data centres. To achieve maximum server performance without throttling, Iceotope’s KUL AI uses an advanced precision cooling solution for faster processing, more accurate results and sustained GPU execution, even for demanding workloads.
KUL AI is highly scalable and proven to achieve up to 4x compaction, handling growing data and model complexity without sacrificing performance.
Its innovative specifications make KUL AI the ideal solution for a range of industries where AI is becoming increasingly essential: from AI research and development centres, HPC labs and cloud service providers (CSPs), to media production and visual effects (VFX) studios, and financial services and quantitative trading firms.
KUL AI offers performance and sustainability benefits
Fitting seamlessly into the KUL family of Iceotope solutions, KUL AI uses Iceotope’s PLC technology which offers several advantages from providing uniform cooling across all heat-generating server components to reducing hotspots and improving overall efficiency. Additionally, PLC eliminates the need for supplementary air cooling, leading to simpler deployments and lower overall energy consumption.
Improving cost-effectiveness and operation efficiency are constant targets for Iceotope. In fact, KUL AI’s advanced thermal management maximises server utilisation, boosting compute density, cutting energy costs and extending hardware lifespan for a lower total cost of ownership (TCO).
Furthermore, KUL AI cuts energy use by up to 40%, and water consumption by 96%, and minimises operational costs while maintaining high thermal efficiency and meeting sustainability targets.
KUL AI offers data centre performance in a range of IT and non-IT environments
Built with scalability and adaptability in mind, KUL AI is deployable in both data centres and, across all Edge IT installations. PLC removes noisy server fans from the cooling process, resulting in near-silent operations and making KUL AI an ideal solution for busy non-IT and populous workspaces which nonetheless demand sustained GPU performance.
Ideal for latency-sensitive Edge deployments and environments with extreme conditions, KUL AI is sealed and protected at the server level, not only ensuring uniform cooling of all components on the GPU motherboard but also rendering it impervious to airborne contaminants and humidity for greater reliability. Crucially, PLC minimises the risk of leaks and system damage, making it a safe choice for any critical environment.
“The unprecedented volume of data being generated by new technologies demands a state-of-the-art solution which not only guarantees server performance but delivers on all vectors of efficiency and sustainability,” said Nathan Blom, Co-CEO of Iceotope. “KUL AI is a pioneering product delivering more computational power and rack space. It offers a scalable system for data centres and is adaptable in non-IT environments, enabling AI everywhere.”
