CSC364 Lecture 13

Introduction to Networked, Distributed, and Parallel Computing
Pipeline
(202607)

Transcript

1. CSC 364 Introduction to Networked, Distributed, and Parallel Computing Lecture 13. Pipeline Dr. Javier Gonzalez-Sanchez javiergs@calpoly.edu www.javiergs.info ffi o ce: 14 -227 1
2. Pipeline Pattern 2
3. De inition • Work lows through st ges (thre ds), typic lly connected by blocking queues. a a a a f f 3 a • E ch st ge runs concurrently.
4. • A Condition object represents st te predic te th t thre ds c n w it for while holding lock. • E ch Condition is tied to one speci ic lock When thre d c lls w it(): • It rele ses the lock • It goes to sleep a a a a a f a a a a a a a a a a a a 4 a When sign led: • It re- cquires the lock • Execution resumes fter w it() a a Class Condition
5. Main (1/2) 5
6. Main (2/2) 6
7. StageQueue 7
8. Life Cycle 8
9. ReaderStage 9
10. ProcessorStage 10
11. StorageStage 11
12. Example 2 12
13. Main (1/2) 13
14. Main (2/2) 14
15. Generator 15
16. Processor 16
17. Reporter 17
18. Depot 18
19. Case 1 19
21. Case 1 21
23. Case 1 23
24. Life Cycle 24
25. Depot 25
26. Code https://github.com/javiergs/CSC364/ tree/main/09-pipeline 26
27. BlockingQueue Java 27
28. De inition A BlockingQueue is thre d-s fe queue th t utom tic lly h ndles: • Mutu l exclusion • W iting when empty a a a a a a a a a a a 28 f • W iting when full
29. De inition It removes the need to m nu lly write: a a 29 f lock.lock(); while(condition) { await(); } ... signal(); lock.unlock();
30. Methods put(E e) • Inserts n element • If the queue is full → the thre d blocks until sp ce is v il ble t ke() a a a a a a a a a a a a a 30 a • Removes nd returns the he d • If the queue is empty → the thre d blocks until n element is v il ble
31. Solution No busy w iting. No m nu l condition v ri bles. No m nu l sign l. It directly solves the Producer–Consumer problem. a a a a a a a 31 a It helps with Pipeline connection
32. Solution 32
33. Questions 33
34. Lab.
35. CSC 364 Introduction to Introduction to Networked, Distributed, and Parallel Computing Javier Gonzalez-Sanchez, Ph.D. javiergs@calpoly.edu Winter 2026 Copyright. These slides can only be used as study material for the class CSC 364 at Cal Poly. They cannot be distributed or used for another purpose. 35