Editing Network congestion (section)

===TCP/IP congestion avoidance===
The [[TCP congestion avoidance algorithm]] is the primary basis for congestion control on the Internet.<ref>[[Van Jacobson]], [[Michael J. Karels]]. [http://citeseer.ist.psu.edu/484335.html Congestion Avoidance and Control] (1988). ''Proceedings of the Sigcomm '88 Symposium'', vol.18(4): pp.314&ndash;329. Stanford, CA. August, 1988. This paper originated many of the congestion avoidance algorithms used in TCP/IP.</ref><ref>RFC 2001 - TCP Slow Start, Congestion Avoidance, Fast Retransmit, and Fast Recovery Algorithms</ref><ref>RFC 2581 - TCP Congestion Control</ref><ref>RFC 3390 - TCP Increasing TCP's Initial Window</ref><ref>[http://www.eventhelix.com/RealtimeMantra/Networking/TCP_Congestion_Avoidance.pdf TCP Congestion Avoidance Explained via a Sequence Diagram]</ref>

Problems occur when concurrent TCP flows experience [[tail-drop]]s, especially when [[bufferbloat]] is present. This delayed packet loss interferes with TCP's automatic congestion avoidance. All flows that experience this packet loss begin a TCP retrain at the same moment – this is called [[TCP global synchronization]].