Abstract

Application Level Framing (ALF) was proposed by Clark and Tennenhouse as an important design principle for developing high performance applications. ALF relies in part on the ability of applications and protocols to process packets independently one from the other. Thus, performance gains one might expect from the use of ALF are clearly related to performance gains one might expect from applications that can handle and process packets received out-of-sequence, as compared to application that require in-sequence delivery (FTP, TELNET, etc.). In this paper, we examine how the ability to process out-of-sequence packets impacts the efficiency of data transmission. We consider both the impact of application parameters such as the time to process a packet by the application, as well as network parameters such as network transmission delay, network loss rate and flow and congestion control characteristics. The performance measure of interest are total latency, buffer requirements, and jitter. We show, using experimental and simulation results, that out-of-sequence processing is beneficial only for very limited ranges of transmission delays and application processing time. We discuss the impact of this on the architecture of communication systems dedicated to distributed multimedia applications.

Author Keywords: Misordered network transmission; Out-of-sequence processing; Application level framing; Partial ordering; Multimedia applications

Article Outline

1. Introduction

2. Issues with out-of-sequence processing

2.1. Why is data misordered?

2.2. Feasibility with respect to applications

2.3. Related works

3. The performance evaluation environment

3.1. Modelling the network

3.1.1. Modelling losses

3.1.2. Modelling transmission delays

3.2. The experimental environment

3.3. The simulation model

3.4. Input parameters

3.5. Performance metrics

3.5.1. Blocking time

3.5.2. Total time

3.5.3. Gain

3.5.4. Jitter

3.5.5. Buffer occupancy

4. Results and analysis

4.1. Buffer requirements

4.2. Blocking time and jitter

4.3. End-to-end transmission time (latency)

4.4. Impact of the transmission parameters on G

4.4.1. Gain vs. window size

4.4.1.1. The model confirms simulation and experimental results.

4.4.1.2. The maximum gain is limited.

4.4.1.3. Out-of-sequence processing is not very useful for large application processing times.

4.4.2. Gain vs. acknowledgment strategy

4.4.3. Gain vs. transmission rate

4.4.4. Gain vs. transmission delay

4.4.5. Gain vs. packet losses

5. Conclusions

Acknowledgements

References

Vitae