Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore advanced techniques for optimizing CPU utilization in Fork/Join frameworks using ManagedBlocker in this 52-minute Devoxx conference talk. Dive into recursive decomposition algorithms and learn how to implement a reserved caching scheme to reduce wasted CPU cycles. Discover the challenges of blocked threads decreasing CPU utilization and uncover a solution using ManagedBlocker in combination with Fork/Join to maintain optimal parallelism. Follow along as Dr. Heinz Kabutz, author of "The Java Specialists' Newsletter," demonstrates these concepts through practical examples, including Fibonacci series calculations, Big Integer operations, and concurrent hash map implementations. Gain insights into measuring multiplication time, creating reserved entries in shared maps, and leveraging ManagedBlocker for improved performance. Conclude with a live demonstration, homework assignment, and Q&A session to reinforce your understanding of these advanced Java concurrency techniques.
Syllabus
Intro
Fibonacci Series
Big Integer
Paralysis
Hacking Big Integer
Measuring Multiplication Time
Creating a Concurrent Hash Map
Putting Reserved into the Map
ManagedBlocker
Reserved Blocker
Demonstration
Homework
Questions
Taught by
Devoxx