Unix - 2

found an article explaining another linux VM "feature":

The Swap Cache

When swapping pages out to the swap files, Linux avoids writing pages if it does not have to. There are times when a page is both in a swap file and in physical memory. This happens when a page that was swapped out of memory was then brought back into memory when it was again accessed by a process. So long as the page in memory is not written to, the copy in the swap file remains valid.

Linux uses the swap cache to track these pages. The swap cache is a list of page table entries, one per physical page in the system. This is a page table entry for a swapped out page and describes which swap file the page is being held in together with its location in the swap file. If a swap cache entry is non-zero, it represents a page which is being held in a swap file that has not been modified. If the page is subsequently modified (by being written to), its entry is removed from the swap cache.

When Linux needs to swap a physical page out to a swap file it consults the swap cache and, if there is a valid entry for this page, it does not need to write the page out to the swap file. This is because the page in memory has not been modified since it was last read from the swap file.

The entries in the swap cache are page table entries for swapped out pages. They are marked as invalid but contain information which allow Linux to find the right swap file and the right page within that swap file.

isn't that great? linux swaps out active data from RAM just to cache it? so instead of actually moving data from swap back into RAM, because it is ACTIVE DATA, it keeps it in swap and now has to keep it in sync by wasting CPU and disk i/o time. Knowing how bad linux VM in relation to swapping is, it creates a huge problem. consider this scenario: you have a bunch of active data in swap and it is being cached in RAM, meaning your RAM is still utilized, then linux decides to cache some more of filesystem and your system is basically out of swap and out of RAM at the same time. your main process using all this RAM(read: the most important process) is killed and you are screwed.

very unimpressive, linux.

In my work I deal with a lot of unix and unix related things. I wouldn't be doing it if I didn't like it. There are, however, 3 things I really dislike. They are mysql, linux and php.

today is about php. please take a look at the following munin graph.



what do you think that was? that huge drop in memory usage, becoming more consistent etc.. It was me downgrading php 5.2.8 to 5.2.6 with all configs being exactly the same. php 5.2.8 turned out to be a real piece of crap. I did try it on a few webservers with high load and it does NOT perform well. the number of system calls is probably 5-10 times higher, I saw spikes in loadaverage, some pages not loading up fast etc. I recommend staying on 5.2.6 for all critical applications. it is not the first time php developers released crap instead of a stable version.

Man, do I hate linux. I normally don't like to vent off like that but take a look at the following munin graph:

look at how linux, having about 5gb of RAM in filesystem cache, decided to offload as much of active data as possible from RAM to swap IN ORDER TO cache more of filesystem.