Multiprocessor system certainty scheduling and load balancing method

Abstract

The invention discloses a multiprocessor-oriented certainty scheduling and load balancing method. According to the method, a certainty scheduling module, a task message queue memory and a load balancing module are provided. The method can be used for carrying out certainty scheduling and load balancing on tasks in multiprocessor systems so as to prevent memory access conflicts and operation errors caused by process scheduling uncertainty. According to the method, the problem that existing multiprocessor systems are low in task distribution efficiency and do not have certainty is solved; through a manner of setting a process recorder to record and manage process IDs and establishing relation with a memory-isolated backup memory space, the memory access conflicts are avoided, so that certainty scheduling is realized; and meanwhile, the leisure and blockage problems caused by low efficiency of fixed processor distribution methods are optimized, and a public message queue method is adopted to distribute processors according to quantity and priorities of tasks via a distribution algorithm.

Description

technical field [0001] The invention is applied to process deterministic scheduling and processor load balancing of multiprocessor systems. Background technique [0002] Process determinism and load balancing technology are necessary for multi-processor systems and platforms. In order to ensure that all processors will not access memory conflicts when accessing memory, the data in memory will be tampered with and affect the normal operation of the process. When the processor system schedules the process, it must ensure determinism; the so-called determinism means that the system can get a certain result after running the process multiple times, and the result is predictable; when the multi-processor system schedules the process, it needs to follow the The deterministic method is used for scheduling to ensure that the results of the system operation are reliable; determinism can be divided into strong determinism and weak determinism, and there are many uncertain factors in t...

AI Technical Summary

Problems solved by technology

[0003] In recent years, with the urgent growth of people's computing needs, multiprocessor systems have become very common; because multiprocessors have strong hardware computing capabilities, they are widely used in various types of computers, embedded systems and various In mobile devices; however, due to the multiple processors, if the system cannot reasonably allocate tasks during tasks, some processor tasks will be overloaded and some processors will be idle, that is, the load is not balanced; It will reduce the efficiency of the system and there is a waste of hardware. Therefore, on a multi-processor system, load balancing technology is very important. Reasonable application of this technology can make the system obtain an objective performance improvement.

Images