yigitsever/kubernetes-dataset

How were the CSV files generated?

Closed this issue · 3 comments

Hi, could you please tell me how were the CSV files generated? From the filenames (such as "background_traffic_20230319T1802.pcap_Flow.csv") I assume that the data were pcaps originally. Which tools were used to generate the CSV files?

The repository is supplementary to the paper A Kubernetes dataset for misuse detection, I was meaning to write a proper README to reflect that

The flow files were generated by this fork of CICFlowMeter

Will the raw pcaps be released?

The raw .pcaps are around 280GB~ compressed, I can surely host them somewhere if needed but the .pcap generation steps are documented in the paper linked above (and I can clarify anything if needed) so you can follow them to generate .pcaps as well