Hi all,


We are currently testing the OPNFV SFC testcase and we have come across some 
worrying performance issues.



Specifically we are running script [0] to create 2 chains (consisting of 1 VNF 
each) and 2 classifiers.

After the script execution has finished we time how long it takes until we see 
the first NXM_NX_NSP flow in table=11.

I list these times below:



1 Compute node   |   7s

2 Compute nodes | 16s

3 Compute nodes | 26s



This measurements are confirmed when we do the same for script [1]



1 Compute node   |   8s

2 Compute nodes | 17s

3 Compute nodes | 32s



The delays are relatively big, but what we find especially worrisome is that 
the delays seem to grow with the number of Compute nodes almost linearly.

This off course is unacceptable, as it cripples the ability to scale in large 
deployments.



The question is, can these operations be run in parallel, so that we have 
almost constant delays relating to the number of compute nodes?



[0] 
https://gerrit.opnfv.org/gerrit/gitweb?p=sfc.git;a=blob;f=tests/functest/odl-sfc/sfc_tacker.bash;h=690d5f52ef8f4410a15edfd030d984894f158053;hb=refs/heads/master



[1] 
https://gerrit.opnfv.org/gerrit/gitweb?p=sfc.git;a=blob;f=tests/functest/odl-sfc/sfc_change_classi.bash;h=70375ab3be4376811ce90e6c3b12531e3bc3462c;hb=refs/heads/master



Best Regards,

George Paraskevopoulos






George Paraskevopoulos

Software Engineer (SDN/NFV), INTRACOM-TELECOM


 + 30 210 667 7689

  <mailto:[email protected]> [email protected]


 19.7 km Markopoulou Ave 19002 Peania, Athens, Greece





The information in this e-mail message and any attachments are intended only 
for the individual or entity to whom it is addressed and may be confidential. 
If you have received this transmission in error, and you are not an intended 
recipient, be aware that any copying, disclosure, distribution or use of this 
transmission or its contents is prohibited.  INTRACOM TELECOM and the sender 
accept no liability for any loss, disruption or damage to your data or computer 
system that may occur while using data contained in, or transmitted with, this 
email. Views or opinions expressed in this message may be those of the author 
and may not necessarily represent those of INTRACOM TELECOM.





_______________________________________________
sfc-dev mailing list
[email protected]
https://lists.opendaylight.org/mailman/listinfo/sfc-dev

Reply via email to