- Author:
- J. Nagle
Most prior work on congestion in datagram systems focuses on buffer management. We find it illuminating to consider the case of a packet switch with infinite storage. Such a packet switch can never run out of buffers. It can, however, still become congested. The meaning of congestion in an infinite-storage system is explored. We demonstrate the unexpected result that a datagram network with infinite storage, first-in-first-out queuing, at least two packet switches, and a finite packet lifetime will, under overload, drop all packets. By attacking the problem of congestion for the infinite-storage case, we discover new solutions applicable to switches with finite storage.
RFC Downloads
Cited By
Cicco L, Gong Y, Rossi D and Leonardi E (2016). A Control-Theoretic Analysis of Low-Priority Congestion Control Reprioritization under AQM, ACM Transactions on Modeling and Performance Evaluation of Computing Systems, 1:4, (1-33), Online publication date: 21-Sep-2016.- Gao F and Qian H (2016). Efficient, real-world token bucket configuration for residential gateways, IEEE/ACM Transactions on Networking, 24:1, (462-475), Online publication date: 1-Feb-2016.
- Glabbeek R and Höfner P (2015). CCS, Acta Informatica, 52:2-3, (175-205), Online publication date: 1-Apr-2015.
- Gettys J and Nichols K (2012). Bufferbloat, Communications of the ACM, 55:1, (57-65), Online publication date: 1-Jan-2012.
- Gettys J and Nichols K (2011). Bufferbloat: Dark Buffers in the Internet, Queue, 9:11, (40-54), Online publication date: 1-Nov-2011.
- Frantti T and Jutila M (2009). Embedded fuzzy expert system for Adaptive Weighted Fair Queueing, Expert Systems with Applications: An International Journal, 36:8, (11390-11397), Online publication date: 1-Oct-2009.
- Kortebi A, Muscariello L, Oueslati S and Roberts J Evaluating the number of active flows in a scheduler realizing fair statistical bandwidth sharing Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systems, (217-228)
- Kortebi A, Muscariello L, Oueslati S and Roberts J (2005). Evaluating the number of active flows in a scheduler realizing fair statistical bandwidth sharing, ACM SIGMETRICS Performance Evaluation Review, 33:1, (217-228), Online publication date: 6-Jun-2005.
- Kortebi A, Muscariello L, Oueslati S and Roberts J Minimizing the overhead in implementing flow-aware networking Proceedings of the 2005 ACM symposium on Architecture for networking and communications systems, (153-162)
- Menasché D, Figueiredo D and de Souza e Silva E (2005). An evolutionary game-theoretic approach to congestion control, Performance Evaluation, 62:1-4, (295-312), Online publication date: 1-Oct-2005.
- Pan R, Breslau L, Prabhakar B and Shenker S (2003). Approximate fairness through differential dropping, ACM SIGCOMM Computer Communication Review, 33:2, (23-39), Online publication date: 1-Apr-2003.
- Heimlich S Traffic characterization of the NSFNET national backbone Proceedings of the 1990 ACM SIGMETRICS conference on Measurement and modeling of computer systems, (257-258)
- Heimlich S (1990). Traffic characterization of the NSFNET national backbone, ACM SIGMETRICS Performance Evaluation Review, 18:1, (257-258), Online publication date: 1-Apr-1990.
- Mills D The fuzzball Symposium proceedings on Communications architectures and protocols, (115-122)
- Mills D (1988). The fuzzball, ACM SIGCOMM Computer Communication Review, 18:4, (115-122), Online publication date: 1-Aug-1988.
- Mills D and Braun H The NSFNET backbone network Proceedings of the ACM workshop on Frontiers in computer communications technology, (191-196)
- Mills D and Braun H (1987). The NSFNET backbone network, ACM SIGCOMM Computer Communication Review, 17:5, (191-196), Online publication date: 1-Oct-1987.
Recommendations
Study of temporal behaviour of packet loss in packet switches with bursty traffic arrivals
The study of packet loss is of great importance to the design of fast packet switching systems. Fast packet switching is generally accepted as the best technique for designing high-speed computer networks. Due to the high throughput demands and the ...
Localized asynchronous packet scheduling for buffered crossbar switches
ANCS '06: Proceedings of the 2006 ACM/IEEE symposium on Architecture for networking and communications systemsBuffered crossbar switches are a special type of crossbar switches. In such a switch, besides normal input queues and output queues, a small buffer is associated with each crosspoint. Due to the introduction of crosspoint buffers, output and input ...
Emulating output queueing with parallel packet switches
This work employs an approximate Markov chain to analyze that a parallel packet switch (PPS) can emulate a first-come first-served (FCFS) output-queued (OQ) packet switch only operating slower than the external line rate. The PPS comprises multiple ...