IBM has developed a new optical technology that can help train AI models at the “speed of light” while also saving significant energy. IBM says that by applying these new breakthroughs to data centers, the energy saved training a typical AI model is equivalent to the annual energy use of 5,000 US households.
According to IBM, while data centers connect to the outside world using fiber optic cables, copper wires are still the predominant material used inside. These copper wires are responsible for connecting GPU accelerators that spend a lot of time waiting for data from other devices while using energy, which significantly increases operating costs. Commenting on the new optical data transmission technology, Dario Gil, senior vice president and director of research at IBM, said:
As AI creates a need for more power and processing power, data centers must evolve – and generic packaged optics technology could help large-scale data centers better adapt to future demands. With this breakthrough, processors of the future will communicate in the same way fiber optic cables carry data in and out of data centers, ushering in a new era of faster, more sustainable communications that can handle massive AI workloads.
IBM outlines its new co-package optics (CPO) technology prototype in a technical paper . Essentially, by dramatically increasing bandwidth in data centers, GPU downtime can be minimized, accelerating AI processing.
IBM claims the technology can reduce the training time for large language models (LLMs) from three months to three weeks. At the same time, the increased energy efficiency will cut down on energy usage and reduce the costs associated with LLM training.
Long-distance travel can involve heavy traffic, changing weather conditions, and rider fatigue. If you are also dealing with the responsibilities of moving home, such as packing belongings or coordinating accommodation, a long ride may add unnecessary pressure to an already busy schedule.
Tired of Microsoft Teams shortcut error preventing you from opening the app? Follow our expert, step-by-step guide with the latest fixes for instant resolution. Works on Windows, Mac & web – no tech skills needed!
Tired of Microsoft Teams Task Management Sync Error halting your workflow? Follow our proven, step-by-step fixes to resolve sync issues fast and restore seamless task collaboration. No tech expertise needed!
Struggling with Microsoft Teams Wiki Error Formatting? This step-by-step guide reveals proven fixes for common wiki tab issues, ensuring smooth editing and collaboration in Teams. Get back to productive wikis fast!
Struggling with Microsoft Teams installation error on Linux? Discover step-by-step fixes for Ubuntu, Fedora & more. Resolve dependency issues, crashes, and errors quickly with our ultimate guide. Get Teams running smoothly today!
Struggling with Microsoft Teams "Error Page" not loading? Get step-by-step fixes for desktop, web, and mobile. Solve Microsoft Teams Error Page issues quickly and resume seamless teamwork today.
Tired of Microsoft Teams "Error Screenshot" blocking your workflow? Get proven, step-by-step solutions to resolve screenshot errors in Teams instantly and boost productivity. No tech skills needed!
Tired of Microsoft Teams "Error U" User blocking your chats? Get proven, step-by-step fixes to clear cache, reset, and restore seamless collaboration instantly.
Unlock the precise locations of Microsoft Teams registry keys on Windows 11. Step-by-step guide to find, access, and safely tweak them for optimal performance and troubleshooting. Essential for IT pros and Teams enthusiasts.
Tired of Microsoft Teams "Training Error" Video Lag ruining your meetings? Follow our step-by-step guide with the latest fixes for smooth video calls—no more frustration!
