✳flâneur — a map of the web's best reading

Enabling "Maximum Enterprise Utilization" with AI

alessiofanelli.com · 737 words · saved by 1 readers

If you spend enough time reading AI papers and articles, you might have run into the acronym "MFU": Maximum FLOPS Utilization. Each GPU has a theoretical FLOPs capacity, but we often only get 20-40% MFU due to issues like sub-optimal scheduling, memory bottlenecks, etc. (See our podcast episode with Quentin Anthony of Eleuther AI (opens in a new tab) for a deeper technical breakdown) You can calculate it with: Where With the rise of AI copilots and agents in the enterprise, more and more tech leaders should be using a similar framework: what is the "Maximum Enterprise Utilization" (MEU) of my company, and how can I adopt AI to improve it? The "All Available Work" for an enterprise is pretty self explanatory, it's all work tasks that need to be done in your company. "Total Productivity" is the sum of all work throughput from all "Workers", which includes employees, contractors, and now AI agents: AI is the first technology in decades that can increase Total Productivity without requiri

Enabling "Maximum Enterprise Utilization" with AI January 3, 2024 If you spend enough time reading AI papers and articles, you might have run into the acronym “MFU”: Maximum FLOPS Utilization. Each GPU has a theoretical FLOPs capacity, but we often only get 20-40% MFU due to issues like sub-optimal scheduling, memory bottlenecks, etc. (See our podcast episode with Quentin Anthony of Eleuther AI for a deeper technical breakdown) You can calculate it with: M F U = Actual FLOPs Theoretical Peak FLOPs MFU=\frac{\text{Actual FLOPs}}{\text{Theoretical Peak FLOPs}} M F U = Theoretical Peak FLOPs Actu

Explore this link on the map →

related reading