Institute for Systems Research Technical Reports
Permanent URI for this collectionhttp://hdl.handle.net/1903/4376
This archive contains a collection of reports generated by the faculty and students of the Institute for Systems Research (ISR), a permanent, interdisciplinary research unit in the A. James Clark School of Engineering at the University of Maryland. ISR-based projects are conducted through partnerships with industry and government, bringing together faculty and students from multiple academic departments and colleges across the university.
Browse
Search Results
Item Dynamics of TCP Congestion Avoidance with Random Drop and Random Marking Queues(2000) Misra, Archan; Baras, John S.; ISR; CSHCNDevelopment and deployment of newer congestion feedback measures such as RED and ECN provide us a significant opportunity for modifying TCP response to congestion. Effective utilization of such opportunities requires detailed analysis of the behavior of congestion avoidance schemes with such randomized feedback mechanisms.In this dissertation, we consider the behavior of generalized TCP congestion avoidance when subject to randomized congestion feedback, such as RED and ECN. The window distribution of individual flows under a variable packet loss/marking probability is established and studied to demonstrate the desirability of specifying a less drastic reduction in the window size in response to ECN-based congestion feedback.
A fixed-point based analysis is also presented to derive the mean TCP window sizes (and throughputs) and the mean queue occupancy when multiple such generalized TCP flows interact with a single bottleneck queue performing randomized congestion feedback.
Recommendations on the use of memory (use of weighted averages of the past queue occupancy) and on the use of "drop biasing" (minimum separation between consecutive drops) are provided to reduce the variability of the queue occupancy.
Finally, the interaction of TCP congestion avoidance with randomized feedback is related to a framework for global optimization of network costs. Such a relation is used to provide the theory behind the shape of the marking (dropping) functions used in a randomized feedback buffer.
Item Window Distribution of Multiple TCPs with Random Loss Queues(1999) Misra, Archan; Baras, John S.; Ott, Teunis; Baras, John S.; ISR; CSHCNIn this paper, we consider the case of multiple ideal and persistent TCP flows (flows that are assumed to be performing idealized congestion avoidance) interacting with queue management algorithms that perform random drop-based buffer management. Our objective is to determine the stationary congestion window distribution of each of the TCP flows whenthe router port implements algorithms like RED (Random Early Detection)or ERD (Early Random Drop).We first present an analyticaltechnique to obtain the 'mean' queue occupancy and the 'mean' of the individual TCP windows. Armed with this estimate of the means, wethen derive the window distribution of each individual TCPconnection. Extensive simulation experiments indicate that, under a wide variety of operating conditions, our analytical method is quite accurate in predicting the 'mean' as well asthe distributions. The derivation of the individual distributions is based upon a numerical analysis presented which considers the case of a single TCP flow subject to variable state-dependent packet loss.
Item A Simulation Study of Enhanced TCP/IP Gateways for Broadband Internet over Satellite(1999) Karir, Manish; Liu, Mingyan D.; Barrett, Bradley A.; Baras, John S.; Baras, John S.; ISR; CSHCNThe demand for Internet bandwidth has been growing rapidlyover the years and the use of high-bandwidth satellites has been proposed as one possible solution to meet the increasingdemand.However, there are certain performance problems withproviding Internet over satellite due to the nature of TCP/IP protocol suite and the satellite link characteristics. In this paper, we describe a simulation study of an architecture for improving the performance of TCP/IP over satellite links.
On each end of the satellite link, there are gateways that split the TCP connection so that the satellite link is transparent to the end hosts.
The split TCP connection over the satellite segment is then optimized.TCP congestion control is maintained on each segment of the split connection.
We simulated such an architecture in OPNET and present results showing improved throughput over the satellite link.
Item Internet Service via Broadband Satellite Networks(1999) Bharadwaj, Vijay G.; Baras, John S.; Butts, Norman P.; Baras, John S.; ISR; CSHCNThe demand for Internet bandwidth has grown rapidly in the past few years. A new generation of broadband satellite constellations promises to provide high speed Internet connectivity to areas not served by optical fiber, cable or other high speed terrestrial connections. However, using satellitelinks to supply high bandwidth has been difficult due to problems with inefficient performance of the Internet's TCP/IP protocol suite over satellite. We describe an architecture for improving the performance of TCP/IP protocols over heterogeneous network environments, especially networks containing satellite links. The end-to-end connection is split into segments, and the protocol on the satellite segment is optimized for the satellite link characteristics. TCP congestion control mechanisms are maintained on each segment, with some coupling between the segments to produce the effect of end-to-end TCP flow control. We have implemented this design and present results showing that using such gateways can improve throughput for individual connections by a large factor over paths containing a satellite link.The research and scientific content in this material has been published in the Proceedings of the SPIE, vol. 3528, February 1999, 169-180. Item Self-Similar Traffic Models(1999) Ramakrishnan, Pradeep; Baras, John S.; ISR; CSHCNWith the advent of broadband communications characterized by a heterogeneous traffic mix (e.g. video conferencing applications, ftp, browsing the web....), commonly held assumptions of traditional traffic models have been put into question. Essentially the present type of traffic is of a highly bursty nature, which is not captured by the traditional traffic models (e.g. Poisson Process). This has a major impact on the design of a network. New models that characterize this burstiness effect are required for the analysis, design, planning, engineering and congestion management of broad-band networks [1].
Measurements using high-resolution traffic equipments of wide area network traffic have confirmed this particular traffic phenomenon. The features shown by the traffic have been called "self-similar or fractal traffic". Their important properties are stated below [1] :-
Distributions of the actual traffic processes decay more slowly (heavy tailed, e.g. of such a distribution is the Pareto distribution) than exponentially (light tailed e.g. a Poisson distribution). See definition of heavy tail and light tailed distribution in the appendix.
Correlations exhibit a hyperbolic (long range dependence) rather than an exponential (short range dependence) decay.
Traditional traffic models used in queueing analysis assume variations only in limited time scales while long-range dependent or self-similar processes fluctuate over a wide range of time scales. This report tries to present various traffic models that represent these properties and the important parameters that need to be estimated which will hopefully enable the design of an optimum network.