What is the Sun-Ni law?
Sun-ni law is an approach used in parallel processing that attempts to improve performance. This is also called memory acceleration and was designed by professors Xian-He Sun and Lionel M. Ni. This law increases the size of the problem and seeks to find a solution limited only by the amount of memory available. It is a generalization of two other approaches used in a parallel calculation called Aldahla's Law and Gustafson's Law. Because it can be difficult to measure, one of the most famous metrics studied is acceleration. Speedup concerns the implementation of parallel programs running on a certain number of processors and the execution time required by the fastest sequential program to solve this problem. One type of approach to accelerating is to maintain a constant size of the problem, allowing you to increase the number of processors that work on the problem. This is called the Law and is known as the acceleration of solid size.
Thus, the AMADHL Act attempts to reduce the time of implementation using parallel processors and correct the computing workload as a constant. In principle, they try to solve the problem in a smaller and smaller time. Gustafson, on the other hand, also known as the acceleration of fixed time, seeks to get the result in a fixed time and increase the size of the problem and perform other operations to obtain an accurate solution. This applies to problems where there is time restriction, but it is not important to solve them in the shortest possible time.
Access to the boundary of the limited acceleration or Sun-Ni law concerns the size of the memory and how it affects performance. The size of the problem that can be solved is affected by the amount of memory available. Limited physical memory means that more time is devoted to finding out solutions to solve the problem in the framework of the computational architecture. The approach that the Sun-Ni law is, if the time limit is fulfilled by the acceleration of fixed time and there is enough memory space, the problem should be adjustedn so that all available memory can be adequate.
This is what the Sun-Ni law does, and the formula considers memory size and is related to performance. Each processor in the architecture of parallel calculations has a fixed memory and the formula combines the size of the problem with the overall available memory capacity. The memory bounded acceleration set out in the Sun-Ni law is essentially a generalization of firm time and solid size. Since the total memory size increases when the number of processors increases, the Sun-Ni law attempts to use all this memory space more efficiently.