Automatic Performance Prediction of Parallel Programs presents a unified approach to the problem of automatically estimating the performance of parallel computer programs. The author focuses primarily on distributed memory multiprocessor systems, although large portions of the analysis can be applied to shared memory architectures as well.
The author introduces a novel and very practical approach for predicting some of the most important performance parameters of parallel programs, including work distribution, number of transfers, amount of data transferred, network contention, transfer time, computation time and number of cache misses. This approach is based on advanced compiler analysis that carefully examines loop iteration spaces, procedure calls, array subscript expressions, communication patterns, data distributions and optimizing code transformations at the program level; and the most important machine specific parameters including cache characteristics, communication network indices, and benchmark data for computational operations at the machine level.
The material has been fully implemented as part of P3T, which is an integrated automatic performance estimator of the Vienna Fortran Compilation System (VFCS), a state-of-the-art parallelizing compiler for Fortran77, Vienna Fortran and a subset of High Performance Fortran (HPF) programs.
A large number of experiments using realistic HPF and Vienna Fortran code examples demonstrate highly accurate performance estimates, and the ability of the described performance prediction approach to successfully guide both programmer and compiler in parallelizing and optimizing parallel programs.
A graphical user interface is described and displayed that visualizes each program source line together with the corresponding parameter values. P3T uses color-coded performance visualization to immediately identify hot spots in the parallel program. Performance data can be filtered and displayed at various levels of detail. Colors displayed by the graphical user interface are visualized in greyscale.
Automatic Performance Prediction of Parallel Programs also includes coverage of fundamental problems of automatic parallelization for distributed memory multicomputers, a description of the basic parallelization strategy and a large variety of optimizing code transformations as included under VFCS.
Le informazioni nella sezione "Riassunto" possono far riferimento a edizioni diverse di questo titolo.
1 Introduction.- 1.1 Automatic Parallelization for Multiprocessor Systems.- 1.2 Motivation for Performance Prediction.- 1.3 P3T: Parameter based Performance Prediction Tool.- 1.4 Overview.- 2 Model.- 2.1 Introduction.- 2.2 Sequential Programs.- 2.3 Parallel Programs.- 2.4 Basic Parallelization Strategy.- 2.5 Optimizing Compiler Transformations.- 2.6 Using P3T and WF under VFCS.- 2.7 Summary.- 3 Sequential Program Parameters.- 3.1 Introduction.- 3.2 Sequential Program Parameters.- 3.3 Instrumentation.- 3.4 Optimization.- 3.5 Adaptation of Profile Data.- 3.6 Summary.- 4 Parallel Program Parameters.- 4.1 Introduction.- 4.2 Work Distribution.- 4.3 Number of Transfers.- 4.4 Amount of Data Transferred.- 4.5 Transfer Time.- 4.6 Network Contention.- 4.7 Number of Cache Misses.- 4.8 Computation Time.- 4.9 Summary.- 5 Experiments.- 5.1 Introduction.- 5.2 Implementation Status.- 5.3 Estimation Accuracy of P3T.- 5.4 Usefulness of P3T.- 5.5 Graphical User Interface of P3T.- 5.6 Summary.- 6 Related Work.- 6.1 Performance Prediction Techniques.- 6.2 Classification of Performance Estimators.- 7 Conclusions.- 7.1 Contributions.- 7.2 Future Research.- A Appendix.- A.1 Intersection and Volume Computation of Convex n-dimensional Polytopes.- A.2 Notation.- References.
Book by Fahringer Thomas
Le informazioni nella sezione "Su questo libro" possono far riferimento a edizioni diverse di questo titolo.
EUR 21,05 per la spedizione da U.S.A. a Italia
Destinazione, tempi e costiEUR 9,70 per la spedizione da Germania a Italia
Destinazione, tempi e costiDa: Zubal-Books, Since 1961, Cleveland, OH, U.S.A.
Condizione: Good. *Price HAS BEEN REDUCED by 10% until Monday, June 9 (weekend SALE item)* 271 pp., hardcover, ex library, else text and binding clean, tight, and bright. - If you are reading this, this item is actually (physically) in our stock and ready for shipment once ordered. We are not bookjackers. Buyer is responsible for any additional duties, taxes, or fees required by recipient's country. Codice articolo ZB386140
Quantità: 1 disponibili
Da: Ammareal, Morangis, Francia
Hardcover. Condizione: Très bon. Ancien livre de bibliothèque. Edition 1996. Ammareal reverse jusqu'à 15% du prix net de cet article à des organisations caritatives. ENGLISH DESCRIPTION Book Condition: Used, Very good. Former library book. Edition 1996. Ammareal gives back up to 15% of this item's net price to charity organizations. Codice articolo E-505-615
Quantità: 1 disponibili
Da: moluna, Greven, Germania
Gebunden. Condizione: New. Automatic Performance Prediction of Parallel Programs presents a unified approach to the problem of automatically estimating the performance of parallel computer programs. The author focuses primarily on distributed memory multiprocessor systems. Codice articolo 458443867
Quantità: Più di 20 disponibili
Da: Ria Christie Collections, Uxbridge, Regno Unito
Condizione: New. In. Codice articolo ria9780792397083_new
Quantità: Più di 20 disponibili
Da: THE SAINT BOOKSTORE, Southport, Regno Unito
Hardback. Condizione: New. This item is printed on demand. New copy - Usually dispatched within 5-9 working days 734. Codice articolo C9780792397083
Quantità: Più di 20 disponibili
Da: Books Puddle, New York, NY, U.S.A.
Condizione: New. pp. 296. Codice articolo 261789132
Quantità: 4 disponibili
Da: Majestic Books, Hounslow, Regno Unito
Condizione: New. Print on Demand pp. 296 52:B&W 6.14 x 9.21in or 234 x 156mm (Royal 8vo) Case Laminate on White w/Gloss Lam. Codice articolo 7140115
Quantità: 4 disponibili
Da: Lucky's Textbooks, Dallas, TX, U.S.A.
Condizione: New. Codice articolo ABLIING23Feb2416190186134
Quantità: Più di 20 disponibili
Da: Biblios, Frankfurt am main, HESSE, Germania
Condizione: New. PRINT ON DEMAND pp. 296. Codice articolo 181789126
Quantità: 4 disponibili
Da: AHA-BUCH GmbH, Einbeck, Germania
Buch. Condizione: Neu. Neuware - Automatic Performance Prediction of Parallel Programs presents a unified approach to the problem of automatically estimating the performance of parallel computer programs. The author focuses primarily on distributed memory multiprocessor systems, although large portions of the analysis can be applied to shared memory architectures as well. The author introduces a novel and very practical approach for predicting some of the most important performance parameters of parallel programs, including work distribution, number of transfers, amount of data transferred, network contention, transfer time, computation time and number of cache misses. This approach is based on advanced compiler analysis that carefully examines loop iteration spaces, procedure calls, array subscript expressions, communication patterns, data distributions and optimizing code transformations at the program level; and the most important machine specific parameters including cache characteristics, communication network indices, and benchmark data for computational operations at the machine level. The material has been fully implemented as part of P3T, which is an integrated automatic performance estimator of the Vienna Fortran Compilation System (VFCS), a state-of-the-art parallelizing compiler for Fortran77, Vienna Fortran and a subset of High Performance Fortran (HPF) programs. A large number of experiments using realistic HPF and Vienna Fortran code examples demonstrate highly accurate performance estimates, and the ability of the described performance prediction approach to successfully guide both programmer and compiler in parallelizing and optimizing parallel programs. A graphical user interface is described and displayed that visualizes each program source line together with the corresponding parameter values. P3T uses color-coded performance visualization to immediately identify hot spots in the parallel program. Performance data can be filtered and displayed at various levels of detail. Colors displayed by the graphical user interface are visualized in greyscale. Automatic Performance Prediction of Parallel Programs also includes coverage of fundamental problems of automatic parallelization for distributed memory multicomputers, a description of the basic parallelization strategy and a large variety of optimizing code transformations as included under VFCS. Codice articolo 9780792397083
Quantità: 2 disponibili