LogP A Practic Parallel

serve as a basis for the design and analysis of fast, portable parallel algorithms, such as algorithms that can be implemented effectively on a wide variety of current and future parallel machines. If we look at the body of parallel algorithms developed under current parallel models, many are impractical because they exploit artificial factors not present in any reasonable machine, such as zero communication delay or infinite bandwidth. Others are overly specialized because they are tailored to the idiosyncrasies of a single machine, such as a particular interconnect topology. To improve this situation, we need to design algorithms using a model that is realistic enough to capture the most important performance factors in real machines, yet abstract enough to be generally useful and to keep the algorithm analysis tractable. Ideally, producing a better algorithm under the model should yield a better program in practice.