Finally, from the point of view of cost, the amount of memory required to store the directory may itself become a bottleneck as the number of processors increases. Recall that the directory size grows as O(mp), where m is the number of memory blocks and p the number of processors. One solution would be to make the memory block larger (thus reducing m for a given memory size). However, this adds to other overheads such as false sharing, where two processors update distinct data items in a program but the data items happen to lie in the same mem...
- 2. Parallel Programming Platforms
- from Introduction to Parallel Computing, Second Edition
- Publisher: Addison-Wesley Professional
- Released: January 2003
Share this highlighthttp://www.safaribooksonline.com/a/introduction-to-parallel/5187311/