Multicore Programming Workshop - Lecture 1

Posted by Ilya Mirman originally on www.cilk.com on Wed, Jul 15, 2009

We recently teamed up with MIT's Professional Development Program on a 2-day workshop focused on multicore programming.

The first lecture covered:

• The multicore programming challenge
• Shared-memory hardware
• Leading concurrency platforms (Pthreads, OpenMP, TBB, Cilk++)
• Race conditions

Here is the video and slides from the first lecture:

• Welcome slides
• Lecture 1 - The Multicore Software Challenge

For more complete information about compiler optimizations, see our Optimization Notice.

Comments (1)

December 19, 2009 2:05 PM PST Emeritus Prof Koh Thiam Chun Total Points: 255 Registered User

The racing problem in parallel processing and programming can be solved through pipeling strategy. The multicore challenge is to meet the expectation on the performance throughput on real time high performance computing (GRID, CLUSTER and Cloud) on web server and other supercomputer processor and dynamic memory allocation on the system stack or the heap LIFO algorithm if possible with virtualisation concept in high performance computing programming (Parallel Programming Concept: Parallel Computational Fluid Dynamic , Hyper-Threading, Hyper-Tasking, Hyper-Data and parallel sematics with linear sematic in parallel Programming theory and applied conceptual programming and the memory DRAM, RAM and ROM is a highly Scalable hardisk more than Raid 5 now and in the future.

Trackbacks (0)

Leave a comment