1. Parallelisation of Desktop Environments Nasser Giacaman Supervised by Dr Oliver Sinnen Department of Electrical and Computer Engineering, The University of Auckland

2. Overview Introduction to parallel computing Motivation State of the art Objectives Methodology Conclusion

3. What is parallel computing?

4. Sequential computingParallel computing

5. Parallel computing challenges: Lack of a central unifying model Communication mechanisms Machine architectures

6. Parallel computing challenges: Task decomposition

7. Parallel computing challenges: Dependency analysis and Scheduling

8. Background to desktop parallelisation Introduction to parallel computing Motivation The need for desktop parallelisation Additional challenges for desktop parallelisation State of the art Objectives Methodology Conclusion

9. The need for desktop parallelisation ~1980

10. The need for desktop parallelisation ~1985~1980

11. The need for desktop parallelisation ~1990~1985~1980

12. The need for desktop parallelisation ~1995~1990~1985~1980

13. The need for desktop parallelisation

14. Parallelise Sequential program Parallel program

15. other applications will be competing for the processor Additional Challenges for Desktop Apps Non-dedicated system

16. Sequential Parallel Additional Challenges for Desktop Apps Overhead sensitive

17. repetitive computation computationally intense one-off computation interactive Additional Challenges for Desktop Apps Different application types

18. Background to desktop parallelisation Introduction to parallel computing Motivation State of the art Typical desktop applications The parallelisation process and tools Objectives Methodology Conclusion

19. worker Thread Structure of a typical desktop application Event loop Event queue Event handler... GUI Thread execute long task Paint GUI GUI update GUI components KEY Thread execution flow data structure code

20. The current situation Most development of desktop applications use Object Oriented (OO) languages and Class libraries 8 of the 10 most popular languages are OO*, such as Java, C++, VB.NET, Python, C#, Ruby * TPC index www.tiobe.com

21. How to parallelise a desktop application In general, 2 types of parallelism: Data parallelism Task parallelism

22. Parallelising compiler Automatic parallelisation Too specific (only simple loops over array variables) Conservative OpenMP Compiler directives Better, but not object oriented (integer-index based) e.g. #pragma omp parallel for for(int i = 0; i < 25; i++){... } How to parallelise a desktop application Data parallelism

23. Thread libraries (e.g. Qt, Java, Pthreads) Everything manual Coarse grained and high overhead Code restructuring OpenMP sections Again, too coarse grained ThreadWeaver, A good approach... but... Code restructuring How to parallelise a desktop application Task parallelism

24. Objectives Re-think the approach of desktop programming Need to provide parallelism in an OO way For both data and task parallelism Find the common “patterns” & merge with OOP Familiar to developers Focus on Maintainable code Benefiting from parallel hardware with minimal effort

25. Overview Introduction to parallel computing Motivation State of the art Objectives Methodology Conclusion

26. ... hasNext()next() Data Parallelism Image resizing - Sequential version

27. List list = getImages(); Iterator it = list.iterator(); while ( it.hasNext() ) { Image image = it.next() resize( image ); } Data Parallelism Image resizing - Sequential version

28. ... Data Parallelism Image resizing – Typical parallel approach

29. ... hasNext()next()hasNext()next()hasNext()next()hasNext()next() Data Parallelism Image resizing – Java-style Iterator

30. List list = getImages(); Iterator it = list.iterator(); Image *array = new Image[ list.size() ]; while ( it.hasNext() ) { array[i] = it.next(); } #pragma omp parallel for for (int i = 0; i < list.size(); i++) { resize( array[i] ); } Data Parallelism Image resizing - OpenMP

31. ... next( image ) Data Parallelism Image resizing – Our solution next( image )

32. List list = getImages(); ParallelIterator it = new ParallelIterator(list); // each thread does this Image image; while (it.next( image )) { resize( image ); } Data Parallelism Image resizing – Our solution

33. Supports all collection types, even if inherently sequential Scheduling on user's behalf static, dynamic, guided chunk size Reductions Global semantics for break statement Negligible overhead Data Parallelism The Parallel Iterator - Results

34. Task Parallelism class MyClass { // class variables void myMethod() { // task A // task B // task C } class MyClass { // class variables void myMethod() { // call tasks } TASK void taskA(){...} TASK void taskB(){...} TASK void taskC(){...} } class TaskA : Thread { // class variables void run() {...} } class TaskB : Thread { // class variables void run() {...} } class TaskC : Thread { // class variables void run() {...} }

35. Task Parallelism Encapsulate a task within a method Follows good OOP practice Unlike threads, these tasks can share resources Issues: Task completion Dependencies task keyword Like OpenMP's section, but support fine-grained parallelism and dependencies

36. Rethinking desktop application structure Queuing events is inherently sequential Create parallelisable tasks Accessing GUI components through multiple threads OO parallel patterns

37. Conclusions Multicores are here! Additional challenges exist for parallel desktop applications Allow the expression of parallelism in a way developers are already familiar with (OOP) e.g. ParallelIterator Data and Task parallelism not enough, need to look at the desktop application structure

38. Than k you! CPU_0 CPU_1

39. “Welcome to the world of Bob the Builder! With their chorus of "Can we fix it?" "Yes, we can!" Bob and his crew of talking machines teach the value of teamwork, problem solving and achieving something by working together” http://nickjr.co.uk/shows/bob/index.aspx From Bob the Builder website...

40. Additional Challenges for Desktop Apps Unknown target system

41. Structure of a typical desktop application Event loop Event handler Paint GUI Event queue... GUI update GUI Thread Sequential desktop program KEY Thread code execution flow data structure

42. Blah blah blah... paralle lising compil er

43. Related work PARC++ (Parallel C++) & PRESTO COOL (Concurrent Object Oriented Language) & CC++ (Compositional C++) PSTL (Parallel Standard Template Library) & STAPL (Standard Template Adaptive Parallel Library) ThreadWeaver Qt Concurrent java.util.concurrent package in Java 1.5

44. The Parallel Iterator Different scheduling types are allowed, as in OpenMP static dynamic guided Different chunk sizes may also be specified chunk size

45. Additional Features 37, 78, 32, 76 sum = 0 37, 78 my_sum = 115 32, 76 my_sum = 108 sum = 223 Reductions Semantics of break for early traversal termination Local vs Global