Hi Huang, CBench throughput mode does not work well in controller with new OF plugin due to a memory issue currently being investigated by OF plugin project. For latency mode you are getting a little

more details » OpenDaylight Release Mgmt & Release Engineering Needs Access Information Where:

Hello OpenDaylight Community: As many of you have heard, Yaiying Wang and his team from Huawei have volunteered to work on filling some of the gaps that we (ODL) have around Release Management and

Here is the meeting Jan was talking yesterday during performance call. Begin forwarded message:

OK, thanks Hideyuki and Venkat for your comments. We are still holding for controller project and some others to upload the stable artifacts into Nexus so we can build the stable editions. BR/Luis

Hi Luis, Here is the result if configure OSGi setting:dropAllPacketsRpc on. Cbench throughput mode: $ taskset -c 0 cbench -c localhost -p 6633 -m 10000 -l 13 -w 3 -M 1000 -t -i 50 -I 5 -s 10 cbench:

Hi Luis, Thank you for making the list. The following patch should not be merged to the stable branch because it is not a bugfix.

I am running the tests as we speak, will join this afternoon. /Jan From: Luis Gomez <ecelgp@...> Date: Monday, April 21, 2014 at 10:00 AM To: Jan Medved <jmedved@...> Cc: "integration-dev@..."

Hi Huang, I agree the performance isn’t up to par with other controller. Here are the results from last night test run http://pastebin.com/A14NTFKJ Chris From: huangxufu <huangxufu@...> Date:

Hi Jan, We have this and other performance results collected by Chris O’Shea, it would be nice if you could join the performance discussion this afternoon at 5:30PM PST BR/Luis Begin forwarded

Hi, can you please rerun your test with the OSGi setting: dropAllPacketsRpc on This should be in theory

Hello all, Thanks for all your kind reminding. First, I agree with Muthukumaran. I have read these two references before, so i chose cbench and controller running on the same VM. Today, i have worked

Hello all, Thanks for all your kind reminding. First, I agree with Muthukumaran. I have read these two references before, so i chose cbench and controller running on the same VM. Today, i have worked

Hi Chris, >>It'srecommended you use a different vm for cbench and the controller .. as long as it is ensured that networklatency / bandwidth constraints do not skew the measurements While running

Hi Luis, Please do not merge this patch 053087de029ed1a3b636c8b74df1b3f04f437042SenthilMon Mar 31 18:23:16 2014 -0700Modify VTN Coordinator tests to work with the latest changes in Coordinator This

Hi, I have identified the following patches in the Integration repo after Hydrogen release. I kind of assume all of them apply to stable release but let me know in case you see any it should

Hi Venkat, No really reason, it is only that throughput mode seems more aggressive than latency mode and thats why we wanted to start with the second, but please feel free to use both in your setup.

Also please check you are using the right settings for Cbench, specially enabling the reactive forwarding module in the OF

Hi, It's recommended you use a different vm for cbench and the controller. This is because both will put heavy load on the CPU. On 19 Apr 2014, at 10:13 pm, "huangxufu" <huangxufu@...> wrote:

Hi Luis, I have executed some tests using cbench for the ODL (Virtualization Edition) for VTN. All the wiki pages regarding cbench in ODL suggest only the "latency" mode option (for cbench). Is