The Bitlet Model: A Parameterized Analytical Model to Compare PIM and CPU Systems

Currently, data-intensive applications are gaining popularity. Together with this trend, processing-in-memory (PIM)–based systems being given more attention and have become relevant. This article describes an analytical modeling tool called Bitlet that can be used in a parameterized fashion to estimate the performance power/energy of PIM-based system and, thereby, assess affinity workloads for PIM as opposed traditional computing. The uncovers interesting trade-offs between, mainly, computation complexity (cycles required perform through PIM), amount memory PIM, bandwidth, data transfer size. Despite its simplicity, model reveals new insights when applied real-life examples. is demonstrated several synthetic examples then explore influence different parameters on two — IMAGING FloatPIM. Based demonstrations, about combination CPU provided.