Performance analysis of reactive congestion control for ATM networks

Kawahara, Kenji; Oie, Yuji; Murata, Masayuki; Miyahara, Hideo

doi:10.1109/49.382156

Cited by 32 publications

(22 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further research may also incorporate Markov-modulated random processes to model bursty behaviors in more detail [23,13,17] and may investigate more complex models by introducing dependences among RVs.…”

Section: Discussionmentioning

confidence: 99%

Stochastic Performance Prediction for Iterative Algorithms in Distributed Environments

Casanova¹,

Thomason²,

Dongarra³

1999

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

The parallelization of iterative algorithms is an important issue for efficient solution of large numerical problems. Several theoretical results concerning sufficient conditions for, and speed of, convergence of parallel iterative algorithms are available. However, those results usually do not take into account the processor workloads and network communications at the application level. The approach in this paper develops a Markov chain based on random variables which describe aspects of the multiuser, distributed-memory environment and the phases of the algorithm. The performance characterization addresses stochastic characteristics of the algorithmic execution time such as mean values and standard deviations. We present simulation results as well as experimental results over different time periods. The results provide information about the impact of distributed environment and implementation style on long-run, expected execution time characteristics. Academic Press

show abstract

Section: Discussionmentioning

confidence: 99%

Stochastic Performance Prediction for Iterative Algorithms in Distributed Environments

Casanova¹,

Thomason²,

Dongarra³

1999

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…Alternatively, the buffer occupancy at the server can also be used as a measure of the level of congestion in the network. The buffer occupancy can be used to control the transmission rate from the video server [7,8]. Feedback from the server buffer has the advantage of a much smaller delay when compared to the delay in obtaining feedback from the network.…”

Section: Congestion Control At the End Systemmentioning

confidence: 99%

Traffic management of multimedia over ATM networks

Zheng

Atiquzzaman²

1999

IEEE Commun. Mag.

View full text Add to dashboard Cite

ultimedia consists of voice, video, and data in the same application. When multimedia is transmitted over a network, video traffic consumes most of the bandwidth, and also requires a high level of quality of service (QoS) guarantee from the network. In order to reduce the bandwidth requirement, video is therefore compressed before transmission over the network. Video compression is based on removing the temporal and spatial redundancy of information in a video stream.An MPEG compressed video consists of different types of frames (I, B, and P), of varying bit rates. A number of these frames are grouped together to form a group of pictures (GoP). A GoP begins with an I-frame, followed by B-and Pframes. If a GoP includes n frames with (m -1) B-frames between consecutive P-frames, the GoP is denoted MmNn. Compression greatly reduces the bit rate of the multimedia stream. However, compressed video is highly bursty in nature. The bit rate changes abruptly from frame to frame; the Iframe has the highest bit rate among the three frame types, and the B-frame usually has the smallest rate. The bursty nature of the multimedia stream necessitates managing the traffic efficiently when transmitted over an asynchronous transfer mode (ATM) network. This gives rise to a number of traffic management issues related to transmission of multimedia traffic over ATM.Advantages of ATM networks, such as high trunk speed, low bit error rate, flexible service type (bandwidth on demand), and high multiplexing capacity make it very suitable for transporting multimedia with QoS guarantees. Statistical multiplexing of bursty video sources in an ATM network improves the bandwidth utilization of the network and lowers the service cost. Furthermore, the multiplexing of multimedia streams reduces the burstiness and bandwidth requirement of the aggregate traffic. However, coincidence of the peaks of many video streams may result in congestion in the network and degrade the quality of the video. Congestion control is important to guarantee the quality of multimedia. Moreover, in multiplexed transmission, efficient bandwidth allocation among the multiple multimedia streams will improve network resource utilization. A network can be managed much more efficiently if the traffic entering the network is smooth. For transmitting video, the bursty multimedia stream can be shaped to smooth out the multimedia traffic stream. Flow control is used to regulate the traffic rate between the sender and the receiver (e.g., a client/server type of system) so that a fast/slow transmitter will not result in overflow/underflow at the receiver. Note that the video quality at the receiver is degraded for either overflow or underflow.When multimedia is transmitted over a dynamically variable bandwidth channel, such as the available bit rate (ABR) service of ATM networks, depending on the level of network congestion, the amount of network bandwidth available to the multimedia source varies over time. In case of congestion, the bit rate of the source can be scaled down to h...

show abstract

“…Congestion degrades the performance of the network and affects the QoS of the connections. Kawahara [26] proposed and analyzed a reactive congestion control algorithm.…”

Section: Video Over Atm With Vbrmentioning

confidence: 99%

Interactive video over ATM: state of the art

Zheng

Atiquzzaman

2003

Computer Communications

View full text Add to dashboard Cite

Multimedia is an emerging service which integrates voice, video and data in the same service. With the progress made in high speed large capacity multimedia servers, high speed networks, QoS, service guarantees and set top boxes, it is currently possible to cost effectively and efficiently carry multimedia over high speed networks. This paper surveys the progress made, and the future of efficiently carrying multimedia over the emerging Asynchronous Transfer Mode networks. q

show abstract

Performance analysis of reactive congestion control for ATM networks

Cited by 32 publications

References 9 publications

Stochastic Performance Prediction for Iterative Algorithms in Distributed Environments

Stochastic Performance Prediction for Iterative Algorithms in Distributed Environments

Traffic management of multimedia over ATM networks

Interactive video over ATM: state of the art

Contact Info

Product

Resources

About