[ndnSIM] ndnSIM Performance

Alex Afanasyev alexander.afanasyev at UCLA.EDU
Fri May 15 16:21:10 PDT 2015


Hi Hisham,

I hope you’re compiling and running code in optimized mode (./waf configure -d optimized).

In optimized mode, ~Ptr() should just decrement a counter, in debug mode there could be more complex things happening, though I’m not sure why would it take 50% of time.

—
Alex

> On May 2, 2015, at 6:59 PM, Hisham Farahat <hishamfarahat at gmail.com> wrote:
> 
> Hi,
> I'm trying to profile ndnSIM 1.0, because My simulation scenarios are taking too long to finish. This was the results:
> CPU: Intel Sandy Bridge microarchitecture, speed 3800 MHz (estimated)
> Counted CPU_CLK_UNHALTED events (Clock cycles when not halted) with a unit mask of 0x00 (No unit mask) count 100000
> samples  %        app name                 symbol name
> 610808   49.4017  ndn         ns3::Ptr<ns3::ndn::Face>::~Ptr()
> 268401   21.7081  python2.7                /usr/bin/python2.7
> 107633    8.7053  ndn        ns3::Ptr<ns3::Packet>::~Ptr()
> 76617     6.1967  ndn         ns3::Ptr<ns3::Packet const>::~Ptr()
> 23462     1.8976  ndn         ns3::Ptr<ns3::ndn::Interest const>::~Ptr()
> .....
> 
> is this normal? 50% on ~Ptr() ?
> Am i doing something wrong?
> 
> Regards,

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 841 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://www.lists.cs.ucla.edu/pipermail/ndnsim/attachments/20150515/213ce34c/attachment.bin>


More information about the ndnSIM mailing list