In evaluating an algorithm, worst-case analysis can be overly pessimistic. Average-case analysis can be overly optimistic. An intermediate approach shows that an algorithm does well on a broad class of input distributions. E. Koutsoupias and C. Ε½ H. Papadimitriou 1994, in ''Proc. of the 35th IEEE An
On competitive on-line paging with lookahead
β Scribed by Dany Breslauer
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 833 KB
- Volume
- 209
- Category
- Article
- ISSN
- 0304-3975
No coin nor oath required. For personal study only.
β¦ Synopsis
This paper studies two methods for improving the competitive efficiency of on-line paging algorithms: in the first, the on-line algorithm can use more pages; in the second, it is allowed to have a lookahead, or in other words, some partial knowledge of the future. The paper considers a new measure for the lookahead size as well as Young's resource-bounded lookahead and proves that both measures have the attractive property that the competitive efficiency of an on-line algorithm with k extra pages and lookahead 1 depends on k + 1. Hence, under these measures, an on-line algorithm has the same benefit from using an extra page or knowing an extra bit of the fnture.
π SIMILAR VOLUMES
within a small constant factor of log D (i.e., they are provably near-optimal) if r < 1/Klog DL; and (4) we introduce a novel admission control policy that partitions the server bandwidth based on the expected popularities of different request lengths and experimentally demonstrate its benefits comp
We study the average competitive ratio of on-line spanning trees with the same distribution of points in the Euclidean plane. We show a distribution of n points such that the average competitive ratio of on-line spanning trees by any on-line algorithm cannot be less than 4 In n -f . @ 1997 Elsevier
This paper studies auctions in a setting where the di erent bidders arrive at di erent times and the auction mechanism is required to make decisions about each bid as it is received. Such settings occur in computerized auctions of computational resources as well as in other settings. We call such au