Questions for Chapter 7
-
A program is optimized by running parts of it in parallel on a machine
with 4 processors.
Before optimization, the parallelizable part of the program accounted
for 80% of the run time of the program.
What is the overall speedup ratio resulting from this optimization,
disregarding communication overhead?
-
What do UMA and NUMA mean in the context of
multiprocessor architectures.
In addition to expanding the acronyms, you answer should indicate the
significance in terms of memory organization.
-
What do the acronyms SISD, SIMD, MISD, and MIMD mean?
-
If multiple processors have private memory address spaces how can they
communicate?