Skip to content
master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 

DeepMPLS: Fast Analysis of MPLS Configurations Using Deep Learning

This repository contains the dataset used for the paper "DeepMPLS: Fast Analysis of MPLS Configurations Using Deep Learning" publish at the IFIP Networking 2019 conference. We refer to the paper for a full explanation of the methodology used for generating the dataset.

DeepMPLS code

Part of the DeepMPLS code is accessible in the dedicated DeepMPLS repository.

Getting the dataset

The dataset is stored in the git repository using git lfs. Install git lfs on your system first and then clone the repository using:

$ git lfs clone https://github.com/fabgeyer/dataset-networking2019.git

The dataset folder should be around 170M.

Reading the dataset

This dataset is based on the networks from the topology-zoo dataset. The repository contains two types of files:

  • Network description files, stored as a tar.gz archive, containing the topologies and the MPLS rules in XML format. The XML format can be processed using P-Rex.
  • Queries files in compressed JSON format.

The matching between network descriptions and queries files is done via the filenames (see example below).

The dataset/qpred folder corresponds to the Satisfiability and Routing tasks in the paper, and the dataset/cpred folder corresponds to the Partial synthesis task.

Example query in the JSON files

In dataset/cpred/topology-zoo/Arpanet196912.queries.json.gz, the first query is:

{
  "query": "<11> SRI .* UCLA <> 0",
  "query_result": true,
  "network": "s1p49"
}

This query corresponds to the network described by the s1p49/topo.xml and s1p49/routing.xml files from the archive dataset/cpred/topology-zoo/Arpanet196912.xmls.tgz.

Citation

If you use this dataset for your research, please include the following reference in any resulting publication:

@inproceedings{GeyerSchmid_Networking2019,
	author    = {Geyer, Fabien and Schmid, Stefan},
	title     = {{DeepMPLS}: Fast Analysis of MPLS Configurations Using Deep Learning},
	booktitle = {Proceedings of the 18th IFIP Networking Conference},
	year      = {2019},
	month     = mai,
	address   = {Warsaw, Poland},
	doi       = {10.23919/IFIPNetworking.2019.8816842},
}

License

The data in this repository is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

About

Dataset used for IFIP Networking 2019

Topics

Resources

Releases

No releases published

Packages

No packages published