[icnrg] Beyond bufferbloat: End-to-end congestion control cannot avoid latency spikes

[Dirk Kutscher <[e-mail]>]

Hi ICNRG,

this article by Bjørn Teigen on the APNIC blog may be of interest to the community:

https://blog.apnic.net/2022/01/26/beyond-bufferbloat-end-to-end-congestion-control-cannot-avoid-latency-spikes/

It includes links to two interesting surveys:

1. https://www.diva-portal.org/smash/record.jsf?pid=diva2%3A1514537&dswid=1621
2. https://www.semanticscholar.org/paper/A-Comprehensive-Overview-of-TCP-Congestion-Control-Lorincz-Klarin/e2f731a675afd6a14cfeb1dde1f602b12394d7b1


In ICN, we have a few leads that indicate the possibility to do better, for example:

* G. Carofiglio, M. Gallo, L. Muscariello, M. Papalini and Sen Wang, "Optimal multipath congestion control and request forwarding in Information-Centric Networks," 2013 21st IEEE International Conference on Network Protocols (ICNP), 2013, pp. 1-10, doi: 10.1109/ICNP.2013.6733576. https://ieeexplore.ieee.org/document/6733576
* Milad Mahdian, Somaya Arianfar, Jim Gibson, and Dave Oran. 2016. MIRCC: Multipath-aware ICN Rate-based Congestion Control. In <i>Proceedings of the 3rd ACM Conference on Information-Centric Networking</i> (<i>ACM-ICN '16</i>). Association for Computing Machinery, New York, NY, USA, 1–10. DOI:https://doi.org/10.1145/2984356.2984365, https://conferences2.sigcomm.org/acm-icn/2016/proceedings/p1-mahdian.pdf


Maybe time for more experimental work for ACM ICN-2022 (and for 6G of course...)? :-)


Best regards,
Dirk