Method for storage area management with static and dynamic joint

Abstract

The method divides the memory area into two portions: one portion is partitioned into static memory blocks in several sizes defined in advance; the other portion is shared dynamic memory area. In static memory, certain number of blocks is allocated for different memory sizes, and blocks will not be split or mergered. The shared dynamic memory area can be allocated dynamically. It is guaranteed that each memory block in various sizes possesses a minimum quantity in order to meet basic application. Meanwhile, limited resources of memory area are utilized fully. The invention assures service efficiency and utilization rate of memory area, solves conflict between shortage of memory area and low availability factor, as well as avoids complex maintenance in operation. The method is applied to management for each memory, data table and periphery memory area, as well as can manages multiform or/and multiple memory areas.

Description

Technical field: [0001] The invention relates to a management method of a storage area, in particular to the management of various storage areas in an embedded real-time system, and is especially suitable for improving the management efficiency of the storage area and improving the management efficiency of the storage area in a system with limited storage area space and high real-time requirements. utilization rate. Background technique: [0002] The memory management of the computer system usually adopts a dynamic management method. When allocating memory, split the memory block of the required size from the memory area. When returning it, if the adjacent space is also free, the memory block must be merged, and often requires multi-level splitting and merge. This method is flexible and can be used to obtain memory blocks of any size, and the memory utilization rate is high. However, due to the frequent splitting and merging of memory blocks, the efficiency is relatively lo...

AI Technical Summary

Problems solved by technology

[0004] First, the memory utilization rate is very low, and it can only be allocated to several sizes of memory currently configured. If the memory of the corresponding size is used up at the time of application, it will return a failure. Even if there are many free memory blocks of other sizes, the application cannot be completed. required memory

[0005] Second, it is necessary to properly configure the number of memories of various sizes in advance. However, different configurations of products and different application environments have great differences in system memory usage requirements. It is necessary to carefully investigate and analyze the usage requirements in advance, otherwise the configured data is unreasonable. Errors that easily lead to memory allocation failure, affecting the normal operation of the business

[0006] Third, it often leads to different configuration parameters and different versions for different users, which is not easy to use and maintain

The first method, its inefficiency is more obvious for embedded real-time systems, and it is not suitable for the application's demand for efficiency

The generation of memory fragmentation will greatly reduce the utilization rate of memory in embedded systems, resulting in failure to apply for larger-sized memory

At the same time, some abnormal applications may exhaust the storage area, causing other basic applications to be unable to be allocated to the necessary storage area, resulting in system failure

[0008] The low utilization rate of the second method is particularly obvious in embedded systems with limited memory space. If the configuration is slightly improper, it will easily lead to memory application failure, resulting in complex system configuration and inconvenient use and maintenance.

[0009] Both of these methods are aimed at the management of heap memory and are suitable for readable and writable storage areas such as DRAM (Dynamic Random Access Memory). They are not suitable for the management of other types of storage areas, such as non-memory data tables and storage space management.

Images