[Starlink] paper: a first look at starlink performance

[Olivier Bonaventure]

Dave,

> So nice to see olivier's name pop up on this (yay!) open access publication:
> 
> https://dl.acm.org/doi/pdf/10.1145/3517745.3561416
> 
> I would have liked if they had tracked the starlink bandwidth
> distributions over time as I did here:
> https://docs.google.com/document/d/1puRjUVxJ6cCv-rgQ_zn-jWZU9ae0jZbFATLf4PQKblM/edit
> 
> and I wonder if the tools they used for their tests are available
> somewhere(?), so they can be repeated over time, especially as more
> ISL links go up.

All the data is available, including pcaps for all experiments

https://smartdata.polito.it/a-first-look-at-starlink-performance-open-data/

François can give you more details about the QUIC implementations used 
for the measurements if you'd like to reproduce them in your testbed. 
The main benefit of QUIC compared to TCP is that it provides more data 
about packet losses and allowed us to explore packet loss bursts


Olivier