29,376 Hits in 9.4 sec

Construction and Use of Linear Regression Models for Processor Performance Analysis

P.J. Joseph, K. Vaswani, M.J. Thazhuthaveetil
The Twelfth International Symposium on High-Performance Computer Architecture, 2006.  
In order to assist architects in making crucial design decisions, we build linear regression models that relate processor performance to micro-architectural parameters, using simulation based experiments  ...  We obtain good approximate models using an iterative process in which Akaike's information criteria is used to extract a good linear model from a small set of simulations, and limited further simulation  ...  In this paper, we draw from past research in the field of design of experiments and linear model construction and propose an iterative process for constructing accurate regression models of processor performance  ... 
doi:10.1109/hpca.2006.1598116 dblp:conf/hpca/JosephVT06 fatcat:4ly6rkp7wvhkdkistzv527djmm

Methods of inference and learning for performance modeling of parallel applications

Benjamin C. Lee, David M. Brooks, Bronis R. de Supinski, Martin Schulz, Karan Singh, Sally A. McKee
2007 Proceedings of the 12th ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '07  
We construct and compare two classes of effective predictive models: piecewise polynomial regression and artifical neural networks.  ...  Increasing system and algorithmic complexity combined with a growing number of tunable application parameters pose significant challenges for analytical performance modeling.  ...  Acknowledgements Part of this work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract No.  ... 
doi:10.1145/1229428.1229479 dblp:conf/ppopp/LeeBSSSM07 fatcat:a7aj3mprzbe2rnnais5bni53my

Eiger: A framework for the automated synthesis of statistical performance models

Andrew Kerr, Eric Anger, Gilbert Hendry, Sudhakar Yalamanchili
2012 2012 19th International Conference on High Performance Computing  
Consequently, we foresee a need for an automated methodology for the systematic construction of performance models of heterogeneous processors.  ...  As processor architectures continue to evolve to increasingly heterogeneous and asymmetric designs, the construction of accurate performance models of execution time and energy consumption has become increasingly  ...  Sandia National Laboratories is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration  ... 
doi:10.1109/hipc.2012.6507525 dblp:conf/hipc/KerrAHY12 fatcat:qbk3daxzxzb5nimcreoikzga7a

Empirical Performance Models for Java Workloads [chapter]

Pradeep Rao, Kazuaki Murakami
2009 Lecture Notes in Computer Science  
Our models use statistical regression techniques to relate overall Java system performance to various observed microarchitecture events and their interactions.  ...  This paper explores and evaluates the effectiveness of empirical modeling for Java workloads.  ...  The authors wish to thank ISIT/Kyushu (JET Programme) for supporting this research, Prof. Nandy, Prof. Matthew Jacob (IISc) and the anonymous reviewers for their comments.  ... 
doi:10.1007/978-3-642-00454-4_22 fatcat:6qq5lubn25a6bnswbl5vax34xm

Characterizing the Effect of Microarchitecture Design Parameters on Workload Dynamic Behavior

Chang-Burm Cho, Wangyuan Zhang, Tao Li
2007 2007 IEEE 10th International Symposium on Workload Characterization  
We then construct error-bounded linear regression models that relate microarchitecture design parameters to various wavelet coefficients that capture workload dynamics at multiresolution levels.  ...  Therefore understanding the effect of design parameters on workload dynamics at early, microarchitecture exploration stage is crucial for highperformance and complexity-efficient designs.  ...  Number of Simulations The number of simulations required for constructing linear regression models that incorporate all input parameters and their complete interaction is prohibitively large.  ... 
doi:10.1109/iiswc.2007.4362176 dblp:conf/iiswc/ChoZL07 fatcat:gi3ukz3tarcjlhqgqhg2blrbxi

A Predictive Performance Model for Superscalar Processors

P. Joseph, Kapil Vaswani, Matthew Thazhuthaveetil
2006 Microarchitecture (MICRO), Proceedings of the Annual International Symposium on  
We evaluate our model building procedure by constructing non-linear performance models for programs from the SPEC CPU2000 benchmark suite with a microarchitectural design space that consists of 9 key parameters  ...  Our models can potentially replace detailed simulation for common tasks such as the analysis of key microarchitectural trends or searches for optimal processor design points.  ...  In this paper, we propose the use of nonlinear regression modeling techniques to build accurate predictive models for processor performance.  ... 
doi:10.1109/micro.2006.6 dblp:conf/micro/JosephVT06 fatcat:s653vjt32jdlthcfeq4iystbhy

Mechanistic-empirical processor performance modeling for constructing CPI stacks on real hardware

Stijn Eyerman, Kenneth Hoste, Lieven Eeckhout
for software and hardware optimization and analysis.  ...  We build mechanistic-empirical performance models for three commercial processor cores, the Intel Pentium 4, Core 2 and Core i7, using SPEC CPU2000 and CPU2006, and report average prediction errors between  ...  Acknowledgements We thank the reviewers for their constructive and insightful feedback. Stijn Eyerman is supported through a postdoctoral fellowship by the Research Foundation-Flanders (FWO).  ... 
doi:10.1109/ispass.2011.5762738 dblp:conf/ispass/EyermanHE11 fatcat:pvy5y5wwcjf3peqw7embkqigci

Modeling and Analyzing the Effect of Microarchitecture Design Parameters on Microprocessor Soft Error Vulnerability

Chang Burm Cho, Wangyuan Zhang, Tao Li
2008 2008 IEEE International Symposium on Modeling, Analysis and Simulation of Computers and Telecommunication Systems  
This paper explores using predictive models to analyze and forecast the effect of various processor microarchitecture design parameters on reliability and their tradeoffs with performance.  ...  High performance and reliability are essential for microprocessor design.  ...  In [12, 13] , quadratic and linear models were proposed to infer runtime microarchitecture AVF using processor performance variables and resource utilization.  ... 
doi:10.1109/mascot.2008.4770557 fatcat:kkb32g34l5davpp3x5kbopeuxa

Comparing Scalability Prediction Strategies on an SMP of CMPs [chapter]

Karan Singh, Matthew Curtis-Maury, Sally A. McKee, Filip Blagojević, Dimitrios S. Nikolopoulos, Bronis R. de Supinski, Martin Schulz
2010 Lecture Notes in Computer Science  
The ANN approach has advantages, but the simpler regression-based model achieves slightly higher accuracy and performance.  ...  Diminishing performance returns and increasing power consumption of single-threaded processors have made chip multiprocessors (CMPs) an industry imperative.  ...  We construct models using artificial neural networks and linear regression, and compare their advantages.  ... 
doi:10.1007/978-3-642-15277-1_14 fatcat:dcsbavyywbdvhi4izdkkwtoipy

Runtime identification of microprocessor energy saving opportunities

W. L. Bircher, M. Valluri, J. Law, L. K. John
2005 Proceedings of the 2005 international symposium on Low power electronics and design - ISLPED '05  
This paper contributes to the solution of these problems by presenting: linear regression models for power consumption and a detailed study of energy efficiency in a modern out-of-order superscalar microprocessor  ...  These simple (2-input) yet accurate (2.6% error) models provide a valuable tool for identifying opportunities to apply power saving techniques such as clock throttling and dynamic voltage scaling (DVS)  ...  Using this finding as a guide we constructed numerous linear models using regression techniques.  ... 
doi:10.1145/1077603.1077668 dblp:conf/islped/BircherVLJ05 fatcat:4ifu3vsjdfdvfenm5einxis2zy

Runtime identification of microprocessor energy saving opportunities

W.L. Bircher, M. Valluri, J. Law, L.K. John
2005 ISLPED '05. Proceedings of the 2005 International Symposium on Low Power Electronics and Design, 2005.  
This paper contributes to the solution of these problems by presenting: linear regression models for power consumption and a detailed study of energy efficiency in a modern out-of-order superscalar microprocessor  ...  These simple (2-input) yet accurate (2.6% error) models provide a valuable tool for identifying opportunities to apply power saving techniques such as clock throttling and dynamic voltage scaling (DVS)  ...  Using this finding as a guide we constructed numerous linear models using regression techniques.  ... 
doi:10.1109/lpe.2005.195527 fatcat:beg7ymxyyjcojgg7xnsfyovpcy

Fine-Grained Energy Consumption Model of Servers Based on Task Characteristics in Cloud Data Center

Zhou Zhou, Jemal H. Abawajy, Fangmin Li, Zhigang Hu, Morshed U. Chowdhury, Abdulhameed Alelaiwi, Keqin Li
2018 IEEE Access  
To this end, we propose a holistic cloud data center energy consumption model that is based on the principal component analysis and regression methods.  ...  As minimizing energy consumption has become a crucial issue for the efficient operation and management of cloud data centers, an energy consumption model plays an important role in cloud datacenter energy  ...  The help of Maliha Omar is also sincerely appreciated.  ... 
doi:10.1109/access.2017.2732458 fatcat:zupugfowvfhfxpwfwbsdmtfn5u

A Novel Method to Predict Processor Performance by Modeling Different Architecture Parameters

Joseph Issa
2020 Journal of Computer Science  
We also present a detailed timing analysis for each processor sub-component. The model is verified to project performance with less than 5% error margin between projected and measured baseline.  ...  Predicting processor throughput and performance is one of the essential aspects of computer architecture.  ...  A regression method is used to determine the relationship to performance in order to construct the performance projection model.  ... 
doi:10.3844/jcssp.2020.479.492 fatcat:bdzvsyxs2nap7ha2lscq5kemu4

Accurate Mutlicore Processor Power Models for Power-Aware Resource Management

Ibrahim Takouna, Wesam Dawoud, Christoph Meinel
2011 2011 IEEE Ninth International Conference on Dependable, Autonomic and Secure Computing  
However, this challenge rises with using virtualization and increasing number of cores in the processors.  ...  In this paper, we analyze power consumption of a multicore processor; we develop three statistical CPU-Power models based on the number of active cores and average running frequency using a multiple liner  ...  ACKNOWLEDGMENTS This work has been done by utilizing resources of HPI Future SOC-Lab that its industrial partners are Fujitsu and Hewlett Packard.  ... 
doi:10.1109/dasc.2011.85 dblp:conf/dasc/TakounaDM11 fatcat:qazcru3dtverneaj2ux5d7rbbu

Predictive modeling based power estimation for embedded multicore systems

Sriram Sankaran
2016 Proceedings of the ACM International Conference on Computing Frontiers - CF '16  
In particular, we construct a per-core based power model using SPLASH2 benchmarks by leveraging concurrency for multicore systems.  ...  In this work, a statistical approach that models the impact of the individual cores and memory hierarchy on overall power consumed by Chip Multiprocessors is developed using Performance Counters.  ...  Linear Regression Linear Regression is typically used to model the relationship between a dependent variable and multiple independent variables.  ... 
doi:10.1145/2903150.2911714 dblp:conf/cd/Sankaran16 fatcat:laqycfmrmrbzjaj5eti6y6oezi
« Previous Showing results 1 — 15 out of 29,376 results