DocumentCode :
770845
Title :
Tree-lookup for partial sums or: how can I find this stuff quickly?
Author :
Beichl, Isabel ; Sullivan, Francis
Author_Institution :
Comput. & Appl. Math. Lab., US Nat. Inst. of Stand. & Technol., Gaithersburg, MD, USA
Volume :
3
Issue :
1
fYear :
1996
Firstpage :
13
Lastpage :
15
Abstract :
Suppose that you need to maintain a large list of data that is to be searched and modified frequently in the course of a computation. You can almost always arrange for both operations to be done efficiently by storing the data in a structure other than a single array. Although there are several elegant ways to do this, we present just one: a simple tree that doesn´t require a lot of thinking to implement. It can speed up execution tremendously because instead of searching through a list item after item for each update, possibly examining n items, you need only examine O(log n) items per operation. (More advanced methods can reduce this to log*n at the cost of more complex programming. Here, log*n is the number of iterations of log2 that are required to get below 1. For numbers having less than 65K bits, log* is less than 5.)
Keywords :
information retrieval; list processing; table lookup; tree searching; data searching; data structure; iterations; large data list; partial sums; simple tree; tree lookup; Binary trees; Costs; Data structures; Epitaxial growth; Molecular beam epitaxial growth; Monte Carlo methods; Physics computing; Random number generation; Springs;
fLanguage :
English
Journal_Title :
Computational Science & Engineering, IEEE
Publisher :
ieee
ISSN :
1070-9924
Type :
jour
DOI :
10.1109/99.486756
Filename :
486756
Link To Document :
بازگشت