Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the principles and techniques of lock-free programming in this 29-minute conference talk by Song Pan. Delve into the world of multi-core processors and parallel computing, understanding how lock-free programming addresses the challenges of shared memory and data races in concurrent systems. Learn about atomic operations and their role in implementing lock-free data structures, using a lock-free stack as a practical example. Examine potential pitfalls such as the ABA problem and memory reordering, and gain insights into the performance benefits of lock-free programming compared to traditional locking methods. Ideal for programmers and computer scientists interested in advanced concurrent programming techniques for multi-threaded applications.
Syllabus
Intro
Background: Trend
Background: Shared Memory
What is lock free?
Atomic Operations (cont'd)
Lock free stack - Push (cont'd)
Lock free stack - Pop (cont'd)
ABA Problem (cont'd)
Memory Reordering (cont'd)
Performance (cont'd)
Taught by
Churchill CompSci Talks