Program Behavior: Models and MeasurementsThis book is intended for two principle classes of readers: those interested in applying program behavior models in their systems modeling efforts, and those interested in program behavior, per se, for research or study purposes. Chapter 1 motivates program behavior modeling and measurement, and introduces several philosophical distinctions to be used later. Chapter 2 deals with those features of paging and paging systems relevant to later discussions: hardware, paging, algorithms, and a more detailed examination of the properties of locality. Chapter 3 presents several basic models, and chapter 4 elaborates on one of the most important modules - the lRU stack model and its extensions. Chapter 5 covers techniques for program measurement and model validation, including an introduction to spectral analysis. Some of the more advanced of these techniques are applied in chapter 6 to two sophisticated program models. Finally, the various models and measurements methods are summarized in chapter 7. |
Contents
Preface | 5 |
Paging Issues | 18 |
Models for Memory Reference Behavior | 56 |
Copyright | |
6 other sections not shown
Other editions - View all
Common terms and phrases
analysis assume autocovariance function autoregressive Belady's lifetime function computed consider correlation correlogram define demand paging demand paging algorithm denote distance probabilities distance string estimated example exponential external memory extrinsic fault probability fault rate curve fetched frequency geometric distribution independent reference model interreference intrinsic locality Jensen's inequality Lewis and Shedler locality list locality set LRU stack model main memory Markov chain mean lifetime measured memory allocation memory reference parameters Pareto phase transitions Poisson process predict prepaging priority algorithm priority list processor program behavior models program model proposition queue random variable real tasks reference string referenced replaced RLRWM sequence set curve set sizes simple A stack simulation SLRUM spectral spectrum stack algorithm stack distance stack model task stationary statistical strong locality condition Suppose survivor function tail-off task's validation values variance virtual VMIN VSLM weak locality condition window