This is a Tensorflow implementation of FlexEntry as described in our paper:

Minghao Ye, Yang Hu, Junjie Zhang, Zehua Guo, H. Jonathan Chao, "Mitigating Routing Update Overhead for Traffic Engineering by Combining Destination-based Routing with Reinforcement Learning," in IEEE Journal on Selected Areas in Communications, vol. 40, no. 9, pp. 2662-2677, Sept. 2022, doi: 10.1109/JSAC.2022.3191337.

• Please install prerequisites (test with Ubuntu 20.04, Python 3.8.5, Tensorflow v2.2.0, Gurobi 9.1.1, networkx 2.5, tqdm 4.51.0)

• For stage 1, to train a sub-model for a topology, unzip and put the topology file (e.g., Ebone) and the traffic matrix file (e.g., EboneTM) in stage1/data/, then specify the file name in config.py, i.e., topology_file = 'Ebone' and traffic_file = 'TM', and then run

python3 stage1/train.py

• For stage 2, to train a stage 2 model for a topology, unzip and put the topology file (e.g., Ebone) and the traffic matrix file (e.g., EboneTM) in stage2/data/, then specify the file name in config.py, i.e., topology_file = 'Ebone' and traffic_file = 'TM', and then run

python3 stage2/train.py

• In a traffic matrix file, each line belongs to a N*N traffic matrix, where N is the node number of a topology.
• Please refer to stage1/config.py and stage2/config.py for more details about configurations.

• To test the sub-model in stage 1 on a set of test traffic matrices, put the test traffic matrix file (e.g., AbileneTM2) in data/, then specify the file name in config.py, i.e., test_traffic_file = 'TM2', and then run

python3 stage1/test.py

• To test the model in stage 2 on a set of test traffic matrices, put the test traffic matrix file (e.g., AbileneTM2) in data/, then specify the file name in config.py, i.e., test_traffic_file = 'TM2', and put the sub-models in stage2/models, then run

python3 stage2/test.py

Please cite our paper if you find our paper/code is useful for your work.

@ARTICLE{ye2022flexentry, author={Ye, Minghao and Hu, Yang and Zhang, Junjie and Guo, Zehua and Chao, H. Jonathan}, journal={IEEE Journal on Selected Areas in Communications}, title={Mitigating Routing Update Overhead for Traffic Engineering by Combining Destination-Based Routing With Reinforcement Learning}, year={2022}, volume={40}, number={9}, pages={2662-2677}, doi={10.1109/JSAC.2022.3191337}}