On the fractal dimension of computer programs and its application to the prediction of the cache miss ratio

Thiébaut, Dominique

doi:10.1109/12.30852

Cited by 65 publications

(42 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MRRL-NSL yields a warmup that is two orders of magnitude smaller than MRRL, see Table 2 which represents the (geometric) average warmup length reduction. The reduction factor increases with increasing K% values because of the fact that the number of unique references within a set of n computer address references grows sublinear with n (Kobayashi and MacDougall, 1989;Thiébaut, 1989). Note that the speedup of MRRL-NSL over MRRL is obtained without sacrificing accuracy.…”

Section: Methodsmentioning

confidence: 90%

Yet shorter warmup by combining no-state-loss and MRRL for sampled LRU cache simulation

Eeckhout

Bosschere

2006

Journal of Systems and Software

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 90%

Yet shorter warmup by combining no-state-loss and MRRL for sampled LRU cache simulation

Eeckhout

Bosschere

2006

Journal of Systems and Software

View full text Add to dashboard Cite

“…This and other metrics and models were discussed at length in Section 6.2. For example, the fractal model by Thiébaut et al [682,683] tends to produce better locality, because it models the jumps between successive addresses using a heavy-tailed distribution, so most of them are very small. In addition, Phalke and Gopinath suggest a model based on inter-reference gaps for temporal locality [547].…”

Section: Memory Behaviormentioning

confidence: 99%

“…To get a self-similar fractal, the jumps from one address to the next must be scale invariant. This is achieved by using a (power-law) Pareto distribution [682,683]. Thus most of the jumps will be small, leading to nearby addresses.…”

Section: Fractal Modelmentioning

confidence: 99%

“…But models of how applications traverse their address space have also been proposed, and provide interesting insights into program behavior [682,683].…”

Section: What It Ismentioning

confidence: 99%

“…An intriguing approach that combines both spatial and temporal aspects of locality, and also characterizes it by a single number, is to measure the fractal dimension of the access Version 1.0.3, typeset on pattern [711,682]. This is based on the postulate that the memory reference pattern of programs has a self-similar structure, i.e.…”

Section: Fractal Modelmentioning

confidence: 99%

See 2 more Smart Citations

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

197

131

View full text Add to dashboard Cite

Reliable performance evaluations require the use of representative workloads. This is no easy task since modern computer systems and their workloads are complex, with many interrelated attributes and complicated structures. Experts often use sophisticated mathematics to analyze and describe workload models, making these models difficult for practitioners to grasp. This book aims to close this gap by emphasizing the intuition and the reasoning behind the definitions and derivations related to the workload models. It provides numerous examples from real production systems, with hundreds of graphs. Using this book, readers will be able to analyze collected workload data and clean it if necessary, derive statistical models that include skewed marginal distributions and correlations, and consider the need for generative models and feedback from the system. The descriptive statistics techniques covered are also useful for other domains.

show abstract

Workload Modeling for Performance Evaluation

Feitelson

2002

Performance Evaluation of Complex Systems: Techniques and Tools

View full text Add to dashboard Cite

The performance of a computer system depends on the characteristics of the workload it must serve: for example, if work is evenly distributed performance will be better than if it comes in unpredictable bursts that lead to congestion. Thus performance evaluations require the use of representative workloads in order to produce dependable results. This can be achieved by collecting data about real workloads, and creating statistical models that capture their salient features. This survey covers methodologies for doing so. Emphasis is placed on problematic issues such as dealing with correlations between workload parameters and dealing with heavy-tailed distributions and rare events. These considerations lead to the notion of structural modeling, in which the general statistical model of the workload is replaced by a model of the process generating the workload.

show abstract

On the fractal dimension of computer programs and its application to the prediction of the cache miss ratio

Cited by 65 publications

References 23 publications

Yet shorter warmup by combining no-state-loss and MRRL for sampled LRU cache simulation

Yet shorter warmup by combining no-state-loss and MRRL for sampled LRU cache simulation

Workload Modeling for Computer Systems Performance Evaluation

Workload Modeling for Performance Evaluation

Contact Info

Product

Resources

About