Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture delves into the concept of locality, focusing on temporal, spatial, and network aspects. It explores the importance of locality in system design, discussing principles like caching, prefetching, and partitioning to optimize data access. The instructor explains how data placement impacts performance, emphasizing the significance of keeping data close to the computation unit. Various strategies, such as hierarchical locks and server-client models, are presented to minimize data movement and enhance system efficiency. The discussion extends to dynamic graphs, memory allocation policies, and causal analysis for performance optimization in multi-threaded applications.