P. Krishnan, P. M. Long and J. S. Vitter. Adaptive disk spindown via optimal rent-to-buy in probabilistic environments. Algorithmica, 23(1):31-56, 1999.

(Available in Postscript and PDF formats.)


Abstract
In the single rent-to-buy decision problem, without a priori knowledge of the amount of time a resource will be used we need to decide when to buy the resource, given that we can rent the resource for $1 per unit time or buy it once and for all for $c. In this paper we study algorithms that make a sequence of single rent-to-buy decisions, using the assumption that the resource use times are independently drawn from an unknown probability distribution. Our study of this rent-to-buy problem is motivated by important systems applications, specifically, problems arising from deciding when to spindown disks to conserve energy in mobile computers, thread blocking decisions during lock acquisition in multiprocessor applications, and virtual circuit holding times in IP-over-ATM networks. We develop a provably optimal and computationally efficient algorithm for the rent-to-buy problem. We describe the experimental results for the application of our algorithm to one of the motivating systems problems: the question of when to spindown a disk to save power in a mobile computer. Simulations using disk access traces obtained from an HP workstation environment suggest that our algorithm yields significantly improved power/response time performance over the non-adaptive 2-competitive algorithm which is optimal in the worst-case competitive analysis model.