Restriction method of Linux page cache based on active reclaiming
The Linux operating system uses a page cache mechanism to cache files on block devices,and the page cache can be infinitely increased until the physical memory reaches the memory watermark.In this case,when the page cache is used excessively,some atomic memory requests in the kernel may fail,resulting in delayed or even failed.The article modifies the memory management module in the kernel to control the use of page cache,so as to maximize the use of page cache.If the maximum limit is reached,the new page cache application will replace the old pages in the page cache.
万方数据
维普
