Hasegawa / Takagi / Takahashi Performance of Distributed Systems and Integrated Communication Networks

Proceedings of the IFIP WG 7.3 International Conference on the Performance of Distributed Systems and Integrated Communication Networks, Kyoto, Japan, 10-12 September, 1991
1. Auflage 2014
ISBN: 978-1-4832-9846-7
Verlag: Elsevier Science & Techn.
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band Volume 5, 436 Seiten, Web PDF

Reihe: IFIP Transactions C: Communication Systems

This book explores new analytical techniques and tools for the performance evaluation of distributed and integrated computer communication systems. The systems considered are those arising in LAN, MAN, WAN broadband ISDN, and ATM switching. These systems are mathematically modelled and analysed. Analytical results are presented on the basic queueing models such as multi-queue, priority queue, queueing network, queue with bursty input and superposed input, and multi-server queue. These results can be usefully applied for the performance evaluation of all the above systems.

Autoren/Hrsg.

Hasegawa, T.

Takagi, H.

Takahashi, Y.

Inhaltsverzeichnis

1;Front Cover;1
2;Performance of Distributed Systems and Integrated Communication Networks;4
3;Copyright Page;5
4;Table of Contents;10
5;PREFACE;6
6;COMMITTEES;8
7;Part I: KEYNOTE ADDRESS;14
7.1;Chapter 1. Technological Competition and Cooperation;14
8;Part II: MAN;20
8.1;Chapter 2. TRAFFIC ENGINEERING FOR HIGH SPEED NETWORKS;20
8.1.1;1. INTRODUCTION;20
8.1.2;2. NETWORKING APPLICATONS AND SERVICES;21
8.1.3;3. NETWORKING CONCEPTS;22
8.1.4;4. HIGH SPEED LOCAL AND METROPOLITAN AREA NETWORKS;24
8.1.5;5. BROADBAND - ISDN;25
8.1.6;6. MAN/ B-ISDN INTERCONNECTION;27
8.1.7;7. TRAFFIC CONTROL AND ENGINEERING;28
8.1.8;CONCLUSION;30
8.1.9;References;31
8.2;Chapter 3. Performance Analysis of the CRMA-Protocol in High- Speed Networks;40
8.2.1;Abstract;40
8.2.2;1. THE CRMA-PROTOCOL;40
8.2.3;2. PERFORMANCE MODEL AND DELAY ANALYSIS;44
8.2.4;3. RESULTS;52
8.2.5;4. CONCLUSION AND OUTLOOK;58
8.2.6;Acknowledgement;59
8.2.7;References;59
8.3;Chapter 4. DQDB Modeling: Problem complexity reduction and solution via Markov chains;60
8.3.1;Abstract;60
8.3.2;1. INTRODUCTION;60
8.3.3;2. DQDB PROTOCOL DESCRIPTION;61
8.3.4;3. DQDB MODELING: REDUCING THE COMPLEXITY;61
8.3.5;4. BUSY TRAIN MODELING;62
8.3.6;5.0 NODE MODELING WITH BWB DISABLED;68
8.3.7;6. CONCLUSIONS;69
8.3.8;REFERENCES;70
9;Part III: Network Performance;76
9.1;Chapter 5. An object-oriented approach for quality of service and network performance modeling;76
9.1.1;Abstract;76
9.1.2;1. INTRODUCTION;76
9.1.3;2. DEFINITION OF QOS AND NP;77
9.1.4;3. QOS AND NP PARAMETERS;78
9.1.5;4. LAYERED NETWORKS A N D SERVICES;80
9.1.6;5. THE ORTHOGONAL MODEL;82
9.1.7;6. AN OBJECT-ORIENTED APPROACH OF QOS A N D NP;83
9.1.8;7. AN EXAMPLE;85
9.1.9;8. CONCLUSIONS;89
9.1.10;9. REFERENCES;89
9.2;Chapter 6. Performance of Connection Management Policies for a LAN to Broadband ISDN Gateway;92
9.2.1;Abstract;92
9.2.2;1.0 Introduction;92
9.2.3;2.0 Description of BERGATE;94
9.2.4;3.0 Modelling Approach;96
9.2.5;4.0 ISDN Channel allocation policies;99
9.2.6;5.0 Arrival Process Description;100
9.2.7;6.0 Measured Characteristics;103
9.2.8;7.0 Conclusions and Directions for Further Work;109
9.2.9;Acknowledgments;110
9.2.10;Abbreviations;110
9.2.11;References;110
10;Part IV: Multi-queue Models;112
10.1;Chapter 7. A HETEROGENEOUS SCAN SERVICE POLLING MODEL WITH SINGLE-MESSAGE BUFFER;112
10.1.1;Abstract;112
10.1.2;1. INTRODUCTION;112
10.1.3;2. THE MODEL;113
10.1.4;3. SOME USEFUL MEASURES FOR THE SYSTEM;115
10.1.5;4. PERFORMANCE MEASURES;119
10.1.6;5. SOME NUMERICAL RESULTS AND CONCLUSIONS;120
10.1.7;6. ACKNOWLEDGEMENT;124
10.1.8;7. REFERENCES;124
10.2;Chapter 8. Analysis of a Priority Queue with Bernoulli Schedules;126
10.2.1;Abstract;126
10.2.2;1. Introduction;126
10.2.3;2. Queueing Model;128
10.2.4;3. Queueing Analysis;129
10.2.5;4. Mean Waiting Times;134
10.2.6;5. Some Comments;140
10.2.7;6. Conclusions;143
10.2.8;References;143
11;Part V: ATM Switching Networks;146
11.1;Chapter 9. Buffered Multistage Interconnection Networks:Simulations Results and Modelling Trials;146
11.1.1;Abstract;146
11.1.2;1. INTRODUCTION;146
11.1.3;2. A PROBABILISTIC MODEL;148
11.1.4;3. APPLICATION OF THE MODEL AND COMPARISON TO THE SIMULATION RESULTS;156
11.1.5;4. CONCLUDING REMARKS;162
11.1.6;5. REFERENCES;164
11.2;Chapter 10. A MODULAR ATM SWITCH FOR BROADBAND ISDN SWITCHING;166
11.2.1;abstract;166
11.2.2;1 INTRODUCTION;166
11.2.3;2 SWITCH ARCHITECTURE;168
11.2.4;3 Modular Architecture of a Large Switch;173
11.2.5;4 PERFORMANCE ANALYSIS;174
11.2.6;5 CONCLUSION;180
11.2.7;References;180
11.3;Chapter 11. Blocking in Asynchronous, Buffered Banyan Networks;182
11.3.1;Abstract;182
11.3.2;1 Introduction;182
11.3.3;2 Model and Notations;183
11.3.4;3 Numerical Examples;200
11.3.5;4 Conclusions;201
11.3.6;References;201
12;Part VI: Modeling Computer Systems;202
12.1;Chapter 12. Performance Analysis of a Processor Sharing Policy with Interactive and Background Jobs;202
12.1.1;Abstract;202
12.1.2;1. INTRODUCTION;202
12.1.3;2. The PSBJ Model;204
12.1.4;3. STABILITY CONDITIONS;205
12.1.5;4. APPROXIMATE ANALYSIS;206
12.1.6;5. ITERATIVE ALGORITHM FOR SOLVING qi's& ITS CONVERGENCE;213
12.1.7;6. NUMERICAL RESULTS;215
12.1.8;7. CONCLUSIONS & FUTURE WORK;218
12.1.9;8. ACKNOWLEDGEMENTS;219
12.1.10;9. REFERENCES;219
13;Part VII: Queueing Networks;222
13.1;Chapter 13. Probabilistic methods for Jackson networks;222
13.1.1;Abstract;222
13.1.2;INTRODUCTION;223
13.1.3;1 Ergodicity conditions for Jackson networks;223
13.1.4;2 Main results;229
13.1.5;3 Geometric construction;230
13.1.6;4 Analytic Lyapounov families;232
13.1.7;5 Lyapounov functions;234
13.1.8;References;235
13.2;Chapter 14. Estimating response time distributions in queueing networks;238
13.2.1;Abstract;238
13.2.2;1. INTRODUCTION;238
13.2.3;2. THE QUEUEING NETWORK MODEL;240
13.2.4;3. CHARACTERIZATION OF THE RESPONSE TIME;241
13.2.5;4. APPLICATION OF UNIFORMIZATION;243
13.2.6;5. MARKOV MONTE CARLO SIMULATION;245
13.2.7;6. STRATIFIED SAMPLING PLAN;246
13.2.8;7. METHOD OF STRATIFICATION;247
13.2.9;8. SAMPLING OF CR USING IMPORTANCE SAMPLING;249
13.2.10;9. NUMERICAL EXAMPLE;252
13.2.11;10. CONCLUSION;252
13.2.12;11. REFERENCES;257
13.3;Chapter 15. Optimal Deadlock Free Buffer Allocation in Multiple Chain Blocking Networks of Queues;258
13.3.1;Abstract;258
13.3.2;1 Introduction;258
13.3.3;2 Model Description;260
13.3.4;3 Conditions For Deadlock Freedom;261
13.3.5;4 Deadlock Free Buffer Allocation Algorithm;263
13.3.6;5 Cycle Finding Algorithm;266
13.3.7;6 Examples;269
13.3.8;7 Conclusions;274
13.3.9;References;274
13.3.10;Appendix;275
13.4;Chapter 16. Asymptotic analysis of closed queueing networks with bottlenecks;278
13.4.1;Abstract;278
13.4.2;1 Introduction;278
13.4.3;2 Networks with a Fixed Number of Stations;279
13.4.4;3 Several Large Groups of LQD Stations;283
13.4.5;4 Numerical Results;286
13.4.6;5 Generalization to Large Multiple Chain Networks;288
13.4.7;6 Conclusions;291
13.4.8;References;292
14;Part VIII: Performance Evalutaion Tools;294
14.1;Chapter 17. Concurrent Stochastic Simulation: Experiments with EcliPSe;294
14.1.1;Abstract;294
14.1.2;1 Introduction;294
14.1.3;2 Issues motivating design;295
14.1.4;3 The EcliPSe Toolkit;296
14.1.5;4 Empirical Results;299
14.1.6;5 Conclusion;311
14.1.7;References;311
15;Part IX: ATM Modeling and Analysis;314
15.1;Chapter 18. Queueing systems for modelling ATM networks;314
15.1.1;Abstract;314
15.1.2;1. ATM networks;314
15.1.3;2. Models of a bursty arrival process;315
15.1.4;3. Discrete-time queueing systems;316
15.1.5;4. The superposition of arrival processes;322
15.1.6;5. Modelling ATM switch architectures;323
15.1.7;6. Congestion control in an ATM network;325
15.1.8;7. Adaptation layer and transport protocols;326
15.1.9;References;327
15.2;Chapter 19. Analysis and Application of an MMPP/PH/n/m Multi-Server Model ;336
15.2.1;Abstract;336
15.2.2;1 Introduction;336
15.2.3;2 Description of the MMPP/PH/n/m model;338
15.2.4;3 Construction of the generator matrix;340
15.2.5;4 Computation of the steady-state distribution by numerical solution methods;344
15.2.6;5 Performance measures of the model;347
15.2.7;6 Investigation of some MMPP/PH/n/m models;351
15.2.8;7 Conclusion;352
15.2.9;Acknowledgment;352
15.2.10;References;353
15.3;Chapter 20. A Method of Computation of the Consecutive Cell Loss Probability for an Individual Source in Superposed Traffic;356
15.3.1;Abstract;356
15.3.2;1 Introduction;357
15.3.3;2 Presentation of our model;358
15.3.4;3 Patterns of consecutive cell loss;361
15.3.5;4 The dynamics of the system;362
15.3.6;5 Study for a Poisson process;364
15.3.7;6 Study for a N-process;365
15.3.8;7 Some particular cases;367
15.3.9;8 Conclusion;374
15.3.10;References;375
15.4;Chapter 21. Cell Loss Behavior in a Statistical Multiplexer with Bursty Input;376
15.4.1;Abstract;376
15.4.2;1 Introduction;377
15.4.3;2 Queueing model and definitions;378
15.4.4;3 Analysis of the lossy and the non-lossy periods;379
15.4.5;4 Cell loss probability in one lossy period;383
15.4.6;5 Numerical Results;388
15.4.7;6 Conclusion;389
15.4.8;References;390
16;Part X: Queueing Systems;396
16.1;Chapter 22. Exact computation of blocking probabilities in state-dependent multi-facility blocking models;396
16.1.1;Abstract;396
16.1.2;1. INTRODUCTION;396
16.1.3;2. RECURSIVE ALGORITHM;399
16.1.4;3. DYNAMIC SCALING;403
16.1.5;4. CONCLUDING REMARKS;404
16.1.6;REFERENCES;404
16.2;Chapter 23. Space priority mechanisms with bursty traffic;406
16.2.1;Abstract;406
16.2.2;1. INTRODUCTION;406
16.2.3;2. SOURCE CHARACTERIZATION;407
16.2.4;3. SYSTEM WITH PARTIAL BUFFER SHARING MECHANISM;409
16.2.5;4. SYSTEM WITH THE PUSH-OUT MECHANISM;414
16.2.6;5. ASSESSMENT OS SPACE PRIORITY MECHANISMS WITH BURSTY TRAFFIC;418
16.2.7;6. DETAILED STUDY OF THE PARTIAL BUFFER SHARING;422
16.2.8;7. CONCLUSIONS;424
16.2.9;Acknowledgement;424
16.2.10;References;424
16.3;Chapter 24. BUFFER SIZE REQUIREMENTS UNDER LONGEST QUEUE FIRST;426
16.3.1;Abstract;426
16.3.2;1 Introduction;427
16.3.3;2 Model Description and Background Material;428
16.3.4;3 Longest Queue First;430
16.3.5;4 Discussion;436
16.3.6;References;437