Explores optimizing library interactions, functionality challenges, and modularity in modern workloads, emphasizing strong boundaries between systems and instruction-level optimizations.
Explores Hadoop's execution models, fault tolerance, data locality, and scheduling, highlighting the limitations of MapReduce and alternative distributed processing frameworks.
Explores reliable broadcast algorithms, focusing on causal order and uniform reliability, with exercises on failure detector properties and memory optimization.
Explores congestion control challenges, emphasizing the need for robust network systems and performance verification to enhance internet performance and address issues like starvation and low utilization.
Explores the impact of locality on system design, discussing strategies to optimize efficiency and performance through data layout, lock design, and data movement minimization.
Explores the evolution of hardware/software co-design, emphasizing the importance of specialization and the challenges of optimizing performance and energy efficiency.