Instant Messaging Meets Video Conferencing: Studying the Performance of IM Video Calls This repository holds the data associated to the TMA 2023 paper of the same name. Publication Laurenz Grote, Ike Kunze, Constantin Sander, and Klaus Wehrle: Instant Messaging Meets Video Conferencing: Studying the Performance of IM Video Calls. In Proceedings of the IFIP/IEEE Network Traffic Measurement and Analysis Conference (TMA '23), 2023. If you use any portion of our work, please consider citing our publication. @Inproceedings {2023-grote-mvca-fairness, title = {Instant Messaging Meets Video Conferencing: Studying the Performance of IM Video Calls}, year = {2023}, month = {6}, publisher = {IFIP/IEEE}, booktitle = {Proceedings of the Network Traffic Measurement and Analysis Conference (TMA '23)}, author = {Grote, Laurenz and Kunze, Ike and Sander, Constantin and Wehrle, Klaus} } data/: Dataset Holds the data set containing all the data plotted in the paper as well as additional measurements. Data format: All measurements are indexed by index.csv. Every row in this file identifies a measurement by a GUID and contains all relevant metadata (scenario type, available bandwidth...). Statistics on the packet level (throughput, queueing, loss and whether the call broke down during the experiment) are supplemented by index_avgbwdist.csv. You should be able to easily join these files on the GUID column. Where applicable, index_qoe.csv contains the aggregate BRISQUE value for the entire call as described in the paper. The bandwidth profile style (Figure 3) measurements are contained in bwprofiles.csv.gz. After decompression, you get all throughput samples of all measurement calls concatenated together. code/: Code Repository Contains the code used to instrument the test bed. Undocumented and never deployed beyond our very own test bed. The code is not runnable as-is. You need prepare the testbed computers as well as configure IP addresses, hostnames, paths and e-mail addresses within the code base. Beyond here might be dragons.