1 |
Am 19.11.2010 16:36, schrieb Alan McKinnon: |
2 |
> Apparently, though unproven, at 17:18 on Friday 19 November 2010, Nikos |
3 |
> Chantziaras did opine thusly: |
4 |
> |
5 |
>> On 11/19/2010 04:37 AM, Adam Carter wrote: |
6 |
>>> 2.6.38 should contain a ~200 line patch that makes a huge difference to |
7 |
>>> desktop responsiveness under load; |
8 |
>>> "Tests done by Mike show the maximum latency dropping by over ten times |
9 |
>>> and the average latency of the desktop by about 60 times" |
10 |
>>> Ref: |
11 |
>>> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 |
12 |
>>> <http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num |
13 |
>>> =1> |
14 |
>>> |
15 |
>>> And a RedHat dev reckons you can get the same via configuration; |
16 |
>>> Ref: |
17 |
>>> http://www.webupd8.org/2010/11/alternative-to-200-lines-kernel-patch.html |
18 |
>>> |
19 |
>>> I havent tried it yet... |
20 |
>> |
21 |
>> Doesn't this patch group tasks by TTY? |
22 |
> |
23 |
> As I understand it, the kernel patch does group by TTY. Personally I think |
24 |
> that's just one way of doing it and there could be others. So it's more proof- |
25 |
> of-concept than TheOneTrueWay(tm) |
26 |
> |
27 |
[...] |
28 |
> |
29 |
> What *I* would like to see is flash goes into it's own group and gets |
30 |
> throttled. Everything else running under KDE is in a different group and left |
31 |
> to run full speed |
32 |
> |
33 |
|
34 |
Please help me understand what this patch/script actually does. As far |
35 |
as I understand it, it groups processes so that the kernel can schedule |
36 |
more fairly among them. It therefore helps to prevent one group of |
37 |
processes from starving all others. Is that correct? |
38 |
|
39 |
First question: What about heavily multi-threaded applications? Does the |
40 |
kernel already make a similar grouping of threads-per-process as it does |
41 |
with processes-per-cgroup? Asked in a different way: Would it have any |
42 |
effect to put single but heavily multi-threaded process such as Tomcat |
43 |
or Apache with Worker MPM into its own dedicated cgroup? |
44 |
|
45 |
Second question: When I run a server with different services, does it |
46 |
make sense to put all services into different cgroups? For example |
47 |
PostgreSQL in the first, Apache in a second and Cron (and thereby all |
48 |
batch jobs) in a third? This should be easy enough to do by editing the |
49 |
init-scripts. |
50 |
|
51 |
Thanks in advance! |
52 |
Florian Philipp |