| 書目名稱 | Multiprocessing |
| 副標題 | Trade-Offs in Comput |
| 編輯 | Vijay K. Naik |
| 視頻video | http://file.papertrans.cn/642/641095/641095.mp4 |
| 叢書名稱 | The Springer International Series in Engineering and Computer Science |
| 圖書封面 |  |
| 描述 | .Multiprocessing: Trade-Offs in Computation andCommunication. presents an in-depth analysis of several commonlyobserved regular and irregular computations for multiprocessorsystems. This book includes techniques which enable researchers andapplication developers to quantitatively determine the effects ofalgorithm data dependencies on execution time, on communicationrequirements, on processor utilization and on the speedupspossible..Starting with simple, two-dimensional, diamond-shaped directed acyclicgraphs, the analysis is extended to more complex and higherdimensional directed acyclic graphs. The analysis allows for thequantification of the computation and communication costs and theirinterdependencies. The practical significance of these results on theperformance of various data distribution schemes is clearly explained.Using these results, the performance of the parallel computations areformulated in an architecture independent fashion. These formulationsallow for the parameterization of the architecture specitific entitiessuch as the computation and communication rates. This type ofparameterized performance analysis can be used at compile time or atrun-time so as to achieve th |
| 出版日期 | Book 1993 |
| 關鍵詞 | Scala; algorithms; parallel computation; processor; requirements |
| 版次 | 1 |
| doi | https://doi.org/10.1007/978-1-4615-3196-8 |
| isbn_softcover | 978-1-4613-6402-3 |
| isbn_ebook | 978-1-4615-3196-8Series ISSN 0893-3405 |
| issn_series | 0893-3405 |
| copyright | Kluwer Academic Publishers 1993 |
|Archiver|手機版|小黑屋|
派博傳思國際
( 京公網(wǎng)安備110108008328)
GMT+8, 2025-10-21 13:36
發(fā)表于 2025-3-21 16:22:52
