Dear Nandana,

Thanks to contacting us. Each compressed file of the dataset contains 500 samples corresponding to one topology, one routing and one traffic matrix. For each of these samples ( a line of the simulationResults.txt) we provide the 10 attributes described in the GitHub repository for each of the path of the topology. For instance, the nsfnet topology which have 14 nodes, has 14*14 attributes = 1960 columns. Despite no traffic is generated between one node to itself, the parameters are included in the sample. The README provide information in how to access each attribute of the path. Notice that we have an API to provide users with a simple and intuitive way to access the information contained in the associated datasets. Each dataset has associated an API which is linked at the beginning of the description of the dataset in the github repository. For dataset v0, you can found the API [here https://github.com/knowledgedefinednetworking/datanetAPI/tree/dataset_v0]

Best regards

Albert López

On 10/10/20 13:26, Nandana Tiwari wrote:

Hello

Hope you're doing well.

I recently found the Knowledge Defined networking repository of datasets and I'm very interested in using them to conduct link fault recovery research in SDN. I understand that the v0 datasets have 500 rows and I have come across the GitHub repository that explains the datasets however it has not clearly mentioned what all attributes are captured. From the repository, I could only find 11 attributes and the dataset has about 1961 columns. If you could please help provide me with information as to what these columns are, I would find it extremely helpful.

Thank you so much!

-- Kind regards Nandana

---

Kdn-users mailing list Kdn-users@knowledgedefinednetworking.org https://mail.n3cat.upc.edu/cgi-bin/mailman/listinfo/kdn-users