کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
9661951 697521 2005 23 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Design and analysis of asymptotically optimal randomized tree embedding algorithms in static networks
موضوعات مرتبط
مهندسی و علوم پایه مهندسی کامپیوتر شبکه های کامپیوتری و ارتباطات
پیش نمایش صفحه اول مقاله
Design and analysis of asymptotically optimal randomized tree embedding algorithms in static networks
چکیده انگلیسی
The problem of dynamic tree embedding in static networks is studied in this paper. We provide a unified framework for studying the performance of randomized tree embedding algorithms which allow a newly created tree node to take a random walk of short distance to reach a processor nearby. In particular, we propose simple randomized algorithms on several most common and important static networks, including d-dimensional meshes, d-dimensional tori, and hypercubes. It is shown that these algorithms, which have small constant dilation, are asymptotically optimal for embedding healthy trees. Our analysis technique is based on random walks on static networks. Hence, analytical expressions for expected load on all the processors are available.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Performance Evaluation - Volume 60, Issues 1–4, May 2005, Pages 141-163
نویسندگان
,