[Planetlab-devel] The physical memory size and virtual memory size of the slices

Sat May 17 05:47:14 EDT 2008

Hi Faiyaz Ahmed,

Thank you for replying my e-mail.
I should have explained the situation more... I’m so sorry.

 >It sounds like you have monitored the memory footprints (virtual and
physical) and CPU utilization of slices running on various nodes using
CoTop.

yes, I had measured these with cotop. I found as follows:

cotop - 09:05:18 up 114 days, 12:39,  0 users,  load average: 5.23,  
5.00, 4.90
Tasks:   7 total,   1 running,   3 sleeping,   1 stopped,   2 zombie
Cpu(s): 45.5% us, 32.0% sy,  0.1% ni, 21.2% id,  0.1% wa,  0.6% hi,   
0.5% si
Mem:   3633548k total,  3219840k used,   413708k free,   143936k buffers
Swap:  1048568k total,       56k used,  1048512k free,  1566876k cached

CTX  TX1 TX15  RX1 RX15 #PR PMEMMB VMEMMB %CPU %MEM NAME
709    0    0    0    0   4   42.5  364.4 27.2  1.2 uzh_lft
   0    0    0    0    0  46   58.6  262.0 13.6  1.6 root
812  294  108   84   50  15   48.8  193.1  7.8  1.4 neclabs_neclc
772    0    0    0    0   7    2.1    4.4  3.9  0.1 rnp_land

736    0    0    0    0   1  187.3 1354.0  3.9  5.3 mpisws_freepastry

530    1    0   12    1   7   13.4   26.7  0.0  0.4 tut_playground
795    0    0    0    0   3    2.8    7.0  0.0  0.1 fokusple_iptel
836    0    0    1    1   7   30.8   46.2  0.0  0.9  
arizona_stork_install
664    0    0    0    0   4    3.0    8.0  0.0  0.1 uci_dandelion
518  469  175    2  159  16   18.4   44.0  0.0  0.5 hplabs_oasis
520    1    3    0    4  34   28.0   78.2  0.0  0.8 princeton_coblitz
587  187   24  187   24   4    4.3   14.1  0.0  0.1 cernettsinghua_xyy
526    1   10    0    3  35   27.4   67.1  0.0  0.8 princeton_codeen
687   11   14   11   13  41  133.7  192.9  0.0  3.8  
princeton_coblitztest
753    4    4    4    5   6   31.9   61.6  0.0  0.9 ufl_ipop
750    6    6    4    6   6   31.9   62.0  0.0  0.9 ufl_ipopvivaldi
766    6    7    5    6   6   32.9   64.8  0.0  0.9 ufl_gvfs
827    0    0    0    0   1    1.6    4.1  0.0  0.0 northwestern_neutro
507    1    0    0    0  12   59.6  145.0  0.0  1.7 pl_netflow

mpisws_freepastry has a very large virtual memory size, would you  
please explain this?
its size has never been changed for 3 days. The node name is  
pl1.planetlab.ics.tut.ac.jp.

Shall I ask something? Sometimes, free memory size on nodes is about  
15-17 Mbyte.
It’s all right to run some programs?

Best,
Chung-Han

----------------------------------------------------------
Chung-Han Lee
Distributed System Lab.
TOYOHASHI UNIVERSITY OF TECHNOLOGY
1-1 Hibarigaoka tempakucho Toyohashi 441-8580 JAPAN
E-mail： lch2244 at gmail.com , lch2244 at hotmail.com
----------------------------------------------------------

On 2008/05/17, at 1:09, Faiyaz Ahmed wrote:

> Hi Chung-Han,
>
> I'm sorry I don't understand your question.  Can you please clarify  
> what
> you are investigating?
>
> It sounds like you have monitored the memory footprints (virtual and
> physical) and CPU utilization of slices running on various nodes using
> CoTop.  It sounds like slices don't all use the same amount of  
> memory or
> CPU.
>
> I don't understand "How to decide these sizes on the node? Who has a
> permission these working?".
>
>
> Faiyaz Ahmed
>
> Lee Chung-Han wrote:
>> Hi, everybody!
>>
>> I had measured free memory size and cpu load on nodes, but free  
>> memory
>> size is always less than 100 Mbyte. After I used
>> “CoTop(http://codeen.cs.princeton.edu/cotop/)”, I  
>> monitored physical
>> memory size and virtual memory size of slices on the nodes.
>> The physical memory size and virtual memory size of the slices were  
>> so
>> different.
>> How to decide these sizes on the node? Who has a permission these  
>> working?
>>
>> Best,
>> Chung-Han Lee
>>
>>
>> ----------------------------------------------------------
>> Chung-Han Lee
>> Distributed System Lab.
>> TOYOHASHI UNIVERSITY OF TECHNOLOGY
>> 1-1 Hibarigaoka tempakucho Toyohashi 441-8580 JAPAN
>> E-mail： lch2244 at gmail.com <mailto:lch2244 at gmail.com> ,
>> lch2244 at hotmail.com <mailto:lch2244 at hotmail.com>
>> ----------------------------------------------------------
>>
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Devel mailing list
>> Devel at lists.planet-lab.org
>> https://lists.planet-lab.org/mailman/listinfo/devel
>
> _______________________________________________
> Devel mailing list
> Devel at lists.planet-lab.org
> https://lists.planet-lab.org/mailman/listinfo/devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.planet-lab.org/pipermail/devel/attachments/20080517/17e686bf/attachment-0001.html