On Mon, Dec 6, 2010 at 6:33 AM, Stroller <strol...@stellar.eclipse.co.uk> wrote: > > On 27/11/2010, at 10:22pm, Nikos Chantziaras wrote: >>> ... >>> Does using the ck-sources kernel reduce the necessity to use ionice? >> >> To some degree, yes. > > Tried this late last week. You chose your words perfectly - there is only > some degree of improvement. > > The machine seems much snappier if I return to an interactive session after > leaving it idle for some time, but during DVD rips it is still unresponsive > for tens of seconds at a time. It seems like maybe it responds quicker than > mainline sources, but it's still so slow that it's both unusable and hard to > be sure whether that's the case. > > Will try cgroups this week, perhaps. > > Stroller.
Hi Stroller, As I understand the cgroups thing it only provides fair CPU access to things started in a terminal. (Or the console as long as the console is still running the app I suppose.) How are you starting your desktop? Are you doing that in some way that fits this model? Also, my bad - I did bad calulations when I responded to this earlier suggest a quick look at the way your machine is caching write data. Let me do that again: c2stable ~ # sysctl -a | grep vm.dirty error: "Invalid argument" reading key "fs.binfmt_misc.register" error: permission denied on key 'net.ipv4.route.flush' vm.dirty_background_ratio = 5 vm.dirty_background_bytes = 0 vm.dirty_ratio = 10 vm.dirty_bytes = 0 vm.dirty_writeback_centisecs = 500 vm.dirty_expire_centisecs = 3000 error: permission denied on key 'net.ipv6.route.flush' c2stable ~ # As vm.dirty_ratio is 10% that says the kernel will cache a write up to 10% of memory before it starts pushing the data out to disk. If the machine has 8GB then that's 800MB. (Not 80MB that I stated earlier.) If your disk subsystem writes data at something like 50MB/Sec then that's 800/50 or about 16 seconds to get it all to disk. As the system will start backing off the write somewhere about vm.dirty_background_ratio maybe it's only half of that but that's still 8 seconds. If you are only seeing this unresponsive nature when doing writes you really should play a little bit with these two parameters. I found they did a lot for my machines. Imagine 24GB @ 10%. I was caching 2.4GB of data in memory and writing it out at about 80MB/Sec, taking something like 30 seconds. Miserable! Good luck! Cheers, Mark