The most widely accepted measure to evaluate the performance of a parallel system is speedup. Speedup (Sɴ) is defined as the ratio of the execution time (Tɴ) on a single processor, to the execution time (Tɴ) on N processors. Let’s discuss more performance measuring parameters of the parallel system.
The theoretical maximum speed that can be achieved with a parallel architecture of N identical processors working concurrently on a problem, is N. This is known as the ideal speedup.
In practice, the speedup is much less, since some architectures do not perform to the ideal level owing to conflicts over memory access, communication delays, inefficient algorithms, and mapping of the natural concurrency in a computing problem.
But in some cases, the speedup can be obtained above the ideal speedup, due to anomalies in programming, compilation, architecture usage, etc.
For example, a single processor system may store all its data off-chip, whereas the multiprocessor system may processor system may store all its data off-chip, whereas the multiprocessor system may store all its data-on chip providing an unpredicted increase in performance.
Performance measuring parameters of the parallel system
Another useful measure in evaluating the performance of a parallel system is efficiency (Eɴ) which can be defined as
Efficiency can be interpreted as providing an indication of the average utilization of “N” processors, expressed as a percentage.
Furthermore, this measure allows a uniform comparison of the various speedups obtained from a system containing a different number of processors.
You may like also
- New Technologies for Parallel Processing
- Introduction to Parallel Processing
- Classification of Parallel Architectures
- General Information about High-Level Programming Languages
