Method and system for dynamically adjusting global heap allocation in multi-thread environment

Abstract

The invention relates to a global heap allocation technology in a multi-thread environment, in particular to a method and system for dynamically adjusting global heap allocation in a multi-thread environment. More specifically, the present invention relates to a method and system for dynamically adjusting global heap allocation by monitoring conflicting parameters of the global heap allocation method. The present invention provides a method for dynamically adjusting global heap allocation in a multi-thread environment, comprising: identifying the global heap allocation method in an application program; judging whether the global heap allocation method is a hotspot of multi-thread conflict; and responding If the judgment result is yes, the memory space requested by the global heap allocation method is allocated by using the local stack. According to the method, the inherent global heap allocation method in the program can be dynamically adjusted according to the real-time running state, reducing the lock competition generated on the global heap, and effectively improving the efficiency of resource allocation and resource utilization.

Description

Technical field [0001] The invention relates to a global heap allocation technology in a multithreaded environment. More specifically, the present invention particularly relates to a method and system for dynamically adjusting global heap allocation by monitoring conflicting parameters of the global heap allocation method. Background technique [0002] At present, with the development of computer multi-core technology, multi-threaded programs have been more and more widely used. There are two major problems in a multi-threaded environment. One is resource sharing, that is, multiple threads share the same physical memory. The second is resource synchronization, that is, the multi-threaded environment requires that the physical memory of the same entity be accessed by only one thread at the same time. This shows that resource sharing and resource synchronization are conflicting requirements between each other. Under the premise of ensuring resource sharing, when multiple threads...

AI Technical Summary

Problems solved by technology

This technical solution can indeed alleviate the lock competition to some extent, but each sub-heap is still allocated multiple threads, and there is still the problem of lock competition

More importantly, when dividing sub-heaps and allocating threads, it is impossible to know which threads should be allocated to which sub-heaps to achieve the best performance at runtime, so often the sub-heap Heap1 is idle, while The sub-heap Heap2 is already in a state of lock competition, resulting in waste of memory resources and a large amount of memory fragmentation

Images