Re: Compressed caching improves LTSP memory efficiency

"Nai Xia" <[email protected]>
Newsgroups gmane.linux.terminal-server.devel
Message-ID <[email protected]>
Thanks a lot :)

On Tue, Jun 3, 2008 at 2:35 AM, Michael Shigorin <[email protected]> wrote:
> On Fri, May 30, 2008 at 04:00:50PM +0800, Nai Xia wrote:
>> hi, guys, we would be very happy if you can provide us some
>> feed back or testing numbers or suggestions on compcache in
>> your distribution/system.  We are really in need of them. :)
>
> Got around while being at the terminal; here you are with numbers
> (PIII-500, 64M RAM):
>
> # uname -a
> Linux ws238 2.6.22-led-tc-alt17 #1 Fri May 16 17:39:21 MSD 2008 i686 GNU/Linux
> # uptime
>  23:15:48 up 6 days,  7:58,  1 user,  load average: 0.10, 0.14, 0.10
> # cat /proc/compcache
> DiskSize:          25968 kB
> NumReads:                     90
> NumWrites:                   609
> FailedReads:                   0
> FailedWrites:                  0
> InvalidIO:                     0
> GoodCompress:                 76 %
> VeryGoodCompress:             54 %
> NoCompress:                    0 %
> CurrentPages:                609
> CurrentMem:                  709 kB
> PeakMem:                     709 kB
> # free
>             total       used       free     shared    buffers     cached
> Mem:         62036      55044       6992          0          0      10448
> -/+ buffers/cache:      44596      17440
> Swap:       288100       2836     285264
> # swapon -s
> Filename                                Type            Size    Used    Priority
> /dev/ramzswap0                          partition       25964   2836    255
> /dev/nbd1                               partition       262136  0       -1
>
> with KDE desktop, Seamonkey running currently with only one
> window and a few tabs, psi, xmms and a bunch of konsoles and
> xterms (rather uncluttered compared to usual situation); that's

Good. And do you have performance numbers or empirical results that
can show the difference  between the system with V.S. without
compcache ?
Or in other words, what observation motivated you to include compcache ?
:)

BTW, I am interested in that enhancement patch, where can I find it ?

> all on ALTSP5 server, so only pixmaps count for the most part.
> Locally running processes:
>
> # ps auxww
> USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
> root         1  0.0  0.4   1536   260 ?        Ss   May27   0:09 init [3]
> root         2  0.0  0.0      0     0 ?        S<   May27   0:00 [kthreadd]
> root         3  0.0  0.0      0     0 ?        S<   May27   0:00 [ksoftirqd/0]
> root         4  0.0  0.0      0     0 ?        S<   May27   0:20 [events/0]
> root         5  0.0  0.0      0     0 ?        S<   May27   0:00 [khelper]
> root        27  0.0  0.0      0     0 ?        S<   May27   0:00 [kblockd/0]
> root        28  0.0  0.0      0     0 ?        S<   May27   0:00 [kacpid]
> root        29  0.0  0.0      0     0 ?        S<   May27   0:00 [kacpi_notify]
> root        91  0.0  0.0      0     0 ?        S<   May27   0:00 [cqueue/0]
> root        92  0.0  0.0      0     0 ?        S<   May27   0:00 [kseriod]
> root       108  0.0  0.0      0     0 ?        S    May27   0:00 [pdflush]
> root       109  0.0  0.0      0     0 ?        S    May27   0:01 [pdflush]
> root       110  0.0  0.0      0     0 ?        S<   May27   0:00 [kswapd0]
> root       111  0.0  0.0      0     0 ?        S<   May27   0:00 [aio/0]
> root       532  0.0  0.0      0     0 ?        S<   May27   0:00 [rpciod/0]
> root       625  0.0  0.3   1660   212 ?        S<s  May27   0:00 /sbin/udevd --daemon
> root       629  0.0  0.2   1516   180 ?        Ss   May27   0:41 /sbin/minilogd
> root       774  0.0  0.0      0     0 ?        S<   May27   0:00 [kpsmoused]
> root       781  0.0  0.0      0     0 ?        S<   May27   0:00 [ksuspend_usbd]
> root       783  0.0  0.0      0     0 ?        S<   May27   0:00 [khubd]
> root      1058  0.0  0.0      0     0 ?        S<   May27   0:00 [ata/0]
> root      1059  0.0  0.0      0     0 ?        S<   May27   0:00 [ata_aux]
> root      1765  0.0  0.3   1608   236 ?        Ss   May27   3:10 /usr/sbin/lbuscd
> root      1767  0.0  0.3   2524   228 ?        Ss   May27   0:00 /usr/sbin/ltspfsd -a
> root      1832  0.0  0.0   1516    24 ?        S    May27   0:00 nbd-client 192.168.1.166 9210 /dev/nbd1
> root      2020  0.0  0.3   1700   232 ?        Ss   May27   0:02 /sbin/rpc.idmapd
> rpc       2060  0.0  0.2   1672   180 ?        Ss   May27   0:00 /sbin/portmap
> root      2218  0.0  0.3   1524   220 ?        Ss   May27   0:00 /usr/sbin/acpid
> root      2255  0.0  0.3   3600   208 ?        Ss   May27   0:00 /usr/sbin/sshd
> root      2284  0.0  0.3   1584   220 ?        Ss   May27   0:00 /usr/sbin/ltspinfod -s
> pulse     2289  0.1  5.2   7768  3252 ?        Ss   May27  15:13 /usr/bin/pulseaudio --system -D --use-pid-file --disallow-module-loading -n -F /etc/ltsp/ltsp.pa -L module-native-protocol-tcp auth-ip-acl=192.168.1.166 -L module-esound-protocol-tcp auth-ip-acl=192.168.1.166 -L module-alsa-sink
> root      2304  0.0  0.5   1968   348 tty7     Ss+  May27   0:00 /bin/sh /usr/share/ltsp/screen.d/xdmcp
> root      2305  0.0  0.3   1516   236 tty1     Ss+  May27   0:00 /sbin/mingetty --noclear tty1
> root      2306  0.0  1.0   2480   636 tty2     Ss   May27   0:00 /bin/login
> root      2329  1.1 57.7  44512 35812 ?        Ss   May27 107:14 X -depth 16 -query 192.168.1.166 -terminate -br -fp tcp/192.168.1.166:7100 vt7 :7
> root      2336  0.0  1.7   2536  1112 tty2     S    May27   0:00 -bash
> root      2730  0.0  0.3   2524   248 ?        S    May27   0:01 /usr/sbin/ltspfsd -a
> root      2784  0.0  1.2   2376   784 tty2     R+   23:17   0:00 ps auxww
>
> --
>  ---- WBR, Michael Shigorin <mike-u2l5PoMzF/[email protected]>
>  ------ Linux.Kiev http://www.linux.kiev.ua/
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_____________________________________________________________________
Ltsp-developer mailing list.   To un-subscribe, or change prefs, goto:
      https://lists.sourceforge.net/lists/listinfo/ltsp-developer
For additional LTSP help,   try #ltsp channel on irc.freenode.net
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.