hyPACK-2013 : Mode-1 (Multi-Core Processors) Laboratory : Topics
Topics dealing with all practical and experimental aspects of various programming paradigms
are considered and example programs are made available to the participants in the laboratory session.
The hyPACK-2013 Mode-1 programme is aimed to understand the practical aspects of performance enhancement through software multi-threading on Multi-Core Processors focusing on thread Programming, Performance tuning aspects,
System Overview of Threading, Software tools, and Memory Allocators from application point of view.
Participants will get an opportunity to walk-through and execute
some of the programs designed
for Mode-1 of this
workshop.
To understand scalability and performance of selective scientific and engineering or commercial applications, minor or substantive
modification of
the hyPACK-2013 software programs may be required. Efforts are on to include State-of-the-Art
Multi-Core Processor Systems as well as GPU based Servers in hyPACK-2013 Laboratory in order to understand performance
issues for large scale application kernels.
Mode-1 (Two days): Performance - Multi-Core Processors - Software Threading
-
Multi Core Processors - An Overview of Architecture & programming environment
-
Software Multi-threading & System Overview of threading
-
Parallel Processing - An Overview of Programming (MPI, POSIX Threads, Intel TBB, OpenMP)
-
Tuning & Performance: Code Optimization for Performance; Compiler Switches
-
Tuning & Performance: An Overview of Math Kernel Libraries on Multi-Cores
-
Tuning & Performance: An Overview of Out-of-Core Libraries on Multi-Cores
-
System and Application Benchmarks and Multi-Threaded I/O Suites on Multi-Cores
-
Performance Issues of Intel Thread Building Blocks (TBB): Memory Allocators on
Multi-Cores
-
Performance Issues of OpenMP 3.X & Pthread Programming on Multi-Cores
-
Performance Issues of MPI 2.X & MPI I/O on Multi-Cores
-
Comparative Performance analysis - OpenMP 3.X, Intel TBB, Pthread Programming, Java Concurrent APIs
-
An Overview of Application Level tuning - Intel Software Tools & Open Source Software tools
Laboratory Session Mode-1 (Two days)
-
Programming exercises
for Numerical and Non-Numerical Computations based on MPI, Pthreads, OpenMP, Java Concurrent APIs, &
Mixed programming
-
Numerical Computations (Dense Matrix Computations, Sparse Matrix Computations), Non-Numerical Computations
(Sorting & Search algorithms)
-
Example programs on Compiler Optimizations - Tuning & Performance issues on Multi-Core Processors
-
Tuning & Performance - Math Kernel Libraries on Multi-Cores, Producer & Consumer Models
-
Tuning & Performance - Selective Application Kernels & System Benchmarks on Multi-Core Processors
-
Demonstration of Software tools on Multi-Core Processors (Thread Profiler, Thread Checker, Thread Performance Analyzer, Memory Allocators)
|