nikon/tuned
1
0
Fork 0
Code Activity Actions

- Updated documentation, thanks to Marcela Maslanova!

Browse source
This commit is contained in:
Philip Knirsch 2009-03-25 17:49:48 +01:00
parent 019e5d08cc
commit fde8c26f21
4 changed files with 27 additions and 27 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
doc/DESIGN.txt
View file

@ -48,7 +48,7 @@ tuned:
together are 100 then this will be a balanced and normalized load
- Accumulated weights per tune plugin can be > 100 or < 100. Need to decide
whether to automatically normalize it or not
/* - Tuning aggressivness can be specified from 0 to 100. 0 == no PM, 100 means
/* - Tuning aggressiveness can be specified from 0 to 100. 0 == no PM, 100 means
always fully tune even under full load. Configurable globally for all tuning
plugins and/or individually */
- Tuning modes should follow the system/user settings for power management of
@ -80,7 +80,7 @@ Monitor plugins:
o Input
o Output
o Blocksize
- Should avoid disk/network io to prevent unecessary wakeups
- Should avoid disk/network io to prevent unnecessary wakeups
Tuning plugins:
@ -94,7 +94,7 @@ Tuning plugins:
o Performance
- Each plugin has to define it's own policy for handling the different
levels of aggregated load
- Each plugin has an interface called setTuning(load) where load is the aggreated
- Each plugin has an interface called setTuning(load) where load is the aggregated
load for that tuning plugin modified by the specific or overall level of
tuning aggressiveness
- Several levels of aggressiveness: none, low, medium, high, max
@ -111,7 +111,7 @@ Tuning plugins:
Example:
--------
monitoring plugin for disk io: Montiors either via /proc or other means the IO
monitoring plugin for disk io: Monitors either via /proc or other means the IO
for all disks in the system.
getLoad() returns something like: {"DISK" : {"sda" : {"READ" : 0.24, "WRITE" :

22
doc/README.utils
View file

@ -2,18 +2,18 @@ Systemtap disk and network statistic monitoring tools
=====================================================
The netdevstat and diskdevstat are 2 systemtap tools that allow the user to
collect detailed information about network and disk aktivity of all
applications running on a system. These 2 tools were insipred by powertop,
which shows the wakeups of applications per second.
collect detailed information about network and disk activity of all
applications running on a system. These 2 tools were inspired by powertop,
which shows number of wakeups for every application per second.
The basic idea is to collect statistic about the running applications in a
form that allows a user to identify applications that behave power
inefficient, meaning instead of doing fewer and bigger IO operations they
do more and smaller ones. Current monitoring tools typically only show
the transfer speeds which isn't very meaningful in that context.
The basic idea is to collect statistic about the running applications in
a form that allows a user to identify power greedy applications.
That means f.e. instead of doing fewer and bigger IO operations they
do more and smaller ones. Current monitoring tools show typically only
the transfer speeds, which isn't very meaningful in that context.
To run them you need to have systemtap and kernel-debuginfo installed.
If both are installed a simple
To run those tools, you need to have systemtap installed.
Then you can simply type
netdevstat
@ -21,7 +21,7 @@ or
diskdevstat
will start the scripts. Both can take up to 3 parameters:
and the scripts will start. Both can take up to 3 parameters:
diskdevstat [Update interval] [Total duration] [Display histogram at the end]
netdevstat [Update interval] [Total duration] [Display histogram at the end]

14
doc/TIPS.txt
View file

@ -1,8 +1,8 @@
=== Simple user tips for improving power usage ===
* Use a properly dimensioned system for the job (no need for overpowered systems for simple Desktop use e.g.)
* For servers consolidate services on fewer systems to maximize efficiency of each system
* Enforce turning of machines that are not used (e.g. company policy)
* Unplug and/or turn of peripherals that aren't used (e.g. external USB devices, monitors, printers, scanners)
* Use a properly dimensioned system for the job (no need for overpowered systems for simple Desktop use e.g.).
* For servers consolidate services on fewer systems to maximize efficiency of each system.
* Enforce turning of machines that are not used (e.g. company policy).
* Unplug and/or turn of peripherals that aren't used (e.g. external USB devices, monitors, printers, scanners).
* Turn of unused hardware already in BIOS.
* Disable power hungry features.
* Enable CPU scaling if supported for ondemand CPU governor. DONT use powersave governor, typically uses more power than ondemand
@ -15,7 +15,7 @@
** hdparm -S240 /dev/sda (20m idle to spindown)
** hdparm -B1 /dev/sda (Max powersave mode)
* Make sure writes to hd don't wake it up too quickly:
** Set flusing to once 5 minutes
** Set flushing to once per 5 minutes
** echo "3000" > /proc/sys/vm/dirty_writeback_centisecs
** Enable laptop mode
** echo "5" > /proc/sys/vm/laptop_mode
@ -29,10 +29,10 @@
** xset dpms 0 0 120
=== Simple programmer tips for improving power usage ===
* Avoid unecessary work/computation
* Avoid unnecessary work/computation
* Wake up only when necessary
* Do not actively poll in programs or use short regular timeouts, rather react to events
* If you wake up, do everything at once (race to idle) and as fast as possible
* Use large buffers to avoid frequent disk access. Write one large block at a time
* Don't use [f]sync() if not necessary
* Group timers accross applications if possible (even systems)
* Group timers across applications if possible (even systems)

10
doc/tuned.8
View file

@ -1,4 +1,4 @@
.TH "tuned" "8" "25 Feb 2009" "Phil Knirsch" "adaptive system tuning daemon"
.TH "tuned" "8" "25 Feb 2009" "Phil Knirsch" "Adaptive system tuning daemon"
.SH NAME
tuned \- dynamic adaptive system tuning daemon
.SH SYNOPSIS
@ -7,7 +7,7 @@ tuned \- dynamic adaptive system tuning daemon
\fBtuned\fR is a dynamic adaptive system tuning daemon
that tunes system settings dynamically depending on
usage. It does so by monitoring the usage of several system components
periodically. Based on that information components will then be put into
periodically. Based on that information components will be put into
lower or higher power saving modes to adapt to the current usage. Currently
only ethernet network and ATA harddisk devices are implemented.
.SH OPTIONS
@ -19,9 +19,9 @@ in the foreground without forking at startup.
.BI \-c "\fR, \fP" \--conffile=" conffile"
Specify the name and path of the configuration file, default is \fB/etc/tuned.conf\fR.
.SH BUGS
If you are using tuned together with NetworkManager at the moment NetworkManager
will completely stop and start a managed device when tuned switches power
saving levels. This happens because the network device will loose it's physical
Using NetworkManager and tuned in the same time will completely stop and start NetworkManager.
The inconvenience is done by tuned, which is setting the power saving levels.
This happens because the network device will loose it's physical
link for a very short time when it performs a power saving switch and
NetworkManager can not distinguish this from a connectivity loss. This will be
resolved in future versions of tuned and NM where NM will gain the mechanism to