search

alibaba-edu / High-Precision-Congestion-Control

294 stars 152 forks source link

Regenerate fig 11c from HPCC paper #30

Open sana-mahmood opened 2 years ago

sana-mahmood commented 2 years ago

Hi, I am trying to recreate fig 11c (fat-tree topology with 50% background fbHadoop traffic). I use run.py to generate config files for each scheme. I am able to get expected results for HPCC and DCTCP, but for DCQCN and TIMELY my values are much more inflated than in the paper. I was wondering if there is any optimization done for DCQCN and TIMELY for the experiments in paper that I am missing out in my experiments, like some particular parameter value?

Thanks

rmiao commented 2 years ago

Hi Sana, would you like to share your results for DCQCN and TIMELY with your parameters so that we have better understanding on your settings?

sana-mahmood commented 2 years ago

Hello, sorry for responding quite late here (it slipped through the cracks) but I would really appreciate your help. Here are my config files for DCQCN and TIMELY with their results

DCQCN config: config_dcqcn.txt

TIMELY config: config_timely.txt

FCT Slowdown Results: For timely, timely_win, dcqcn and dcqcn_win, my results have higher values compared to those in the paper

image

Paper Result for comaprison:

image

Thank you for taking the time to answer my question!