Software Group Compiler Technology Task thread and processor

Software Group Compiler Technology Task, thread and processor — Open. MP 3. 0 and beyond Guansong Zhang, IBM Toronto Lab © 2006 IBM Corporation

Compiler technology Overview § The purpose of the talk – Introducing the latest improvement on the Open. MP standard • Task • Still under discussion, don’t take the syntax – Considerations for the future Open. MP development • Thread affinity 2

Compiler technology The changing world § Hardware improvement – Development of the multicore system • Soon we will have more processors than we know how to program with • IBM to Build World's First Cell Broadband Engine Based Supercomputer • Intel: Quad core to turbocharge chips • Terra Soft to Build Cell-Based Super Out of PS 3 Beta Iron § Open. MP Standard – C/C++ and Fortran standard are merged into 2. 5 § Other language committee – C++ memory model: atomic access 3

Compiler technology More changes in the Open. MP world § New players – Microsoft just joined the Open. MP ARB • …, and Visual C++® 2005 supports the full standard. Open. MP is also supported by the Xbox 360™ platform – GCC • The GOMP project is developing an implementation of Open. MP for the C, C++, and Fortran 95 compilers in the GNU Compiler Collection 4

Compiler technology Overview § The purpose of the talk – Introducing the latest improvement on the Open. MP standard • Task • Still under discussion, don’t take the syntax – Considerations for the future Open. MP development • Thread affinity 5

Compiler technology Workshare and task pool § What is a workshare 6

Compiler technology Workshare and task pool (cont. ) § What is a task § Not a workshare – But still “sharing/cooperating” between threads § Comparing with a workshare – Unit can be generated – Unit can wait for another generated unit 7

Compiler technology Task examples § Pointer chasing #pragma omp parallel § Recursive algorithm int fib(int n) { int x, y; { if (n<2) #pragma omp single return n; { #pragma omp taskgroup while(p) { { #pragma omp task common(x) process(p) x=fib(n-1); p=p->next; #pragma omp task common(y) } y=fib(n-2); } } } return x+y; } 8

Compiler technology Task schedule § More flexible scheduling – Can a task be multi-threaded? • When a task is encountered, the thread always go for the new task § Advantage – The idea is to provide one more level of abstraction • • • Task centric view Try to avoid thread starvation Potential cache reuse § Disadvantage – Threadprivate • No threadprivate data – Thread id • HPC users may need thread id to localize data access. – Locks • Locks’ owner becomes confusing 9

Compiler technology Overview § The purpose of the talk – Introducing the latest improvement on the Open. MP standard • Task • Still under discussion, don’t take the syntax – Considerations for the future Open. MP development • Thread affinity 10

Compiler technology Emerging architectures 11

Compiler technology Performance number Stride 1 Stride 2 12

Compiler technology Thread affinity § Nested parallelism • Organize threads to multi levels (This is in previous Open. MP standard already) § Thread grouping • Balancing the number of threads available and the parallelism in the code § Thread mapping • Associate each Open. MP thread to physical/logical processors 13

Compiler technology How to represent a thread group Environment Var Explicit index User interface No touch for user code Simple data type; New internal type, Possible multiple allow centralized changes in the source. thread group programming Modularity (procedure calls, library functions) No support Pass level, CPU array and array size Pass group type var, which may be used as an execution context (MPI) Nested par (thread to thread affinity) Fixed in advance, no dynamic adjustment according to user input Specify number of threads at different levels Specify thread composition Mapping threads Implementation defined Supported, through Virtual CPU numbers Supported, through omp_get_procs() Heterogeneous system No support (? ) Supported, different kinds of CPU with same numbering scheme? Different groups 14 Descriptor handle

Compiler technology What is for the future § Performance is still our goal – “Open. MP is about performance. ” • Quoted from NASA scientists § Open. MP needs to enlarge itself for the broader market – C/C++ will become more interesting – People like to see non numeric programs in Open. MP § Partition Open. MP interface as different layers – TASK, WORKSHARE vs. THREAD vs. PROC – MPI has more than 300 calls, most people will only use 6 -8 – Keep the layered approach while we extending Open. MP ? 15

Compiler technology Summary § Start a parallel region § Split into two nested parallel regions – This is the chance to bind threads to the right processors § Start a task region – For independent works • E. g. game objects § Start a workshare – For computation intensive calculation • 16 E. g. graphic rendering

Compiler technology Q&A 17