Integrated prefetching and caching in single and parallel disk systems

Abstract

We study integrated prefetching and caching in single and parallel disk systems. There exist two very popular approximation algorithms called Aggressive and Conservative for minimizing the total elapsed time in the single disk problem. For D parallel disks, approximation algorithms are known for both the elapsed time and stall time performance measures. In particular, there exists a D-approximation algorithm for the stall time measure that uses D−1 additional memory locations in cache.In the first part of the paper we investigate approximation algorithms for the single disk problem. We give a refined analysis of the Aggressive algorithm, showing that the original analysis was too pessimistic. We prove that our new bound is tight. Additionally we present a new family of prefetching and caching strategies and give algorithms that perform better than Aggressive and Conservative.In the second part of the paper we investigate the problem of minimizing stall time in parallel disk systems. We present a polynomial time algorithm for computing a prefetching/ caching schedule whose stall time is bounded by that of an optimal solution. The schedule uses at most 3(D − 1) extra memory locations in cache. This is the first polynomial time algorithm for computing schedules with a minimum stall time. Our algorithm is based on the linear programming approach of [1]. However, in order to achieve minimum stall times, we introduce the new concept of synchronized schedules in which fetches on the D disks are performed completely in parallel. * Work supported by the Deutsche Forschungsgemeinschaft, project project AL 464/3-1, and by the EU, projects APPOL and APPOL II.

Finding Total Unimodularity in Optimization Problems Solved by Linear Programs

“…While the real life problem is on-line, and has been extensively studied by Cao et al [7], the offline problem has first been solved in 1998 [3], by the use of a linear program, for which it was shown that it always has an optimal integer solution, while not being totally unimodular. Later in 2000 [2], a polynomial time algorithm was given modeling the problem as a multi commodity flow with some postprocessing. Formally the problem can be defined as follows.…”

“…As a result, the tall/small job scheduling problem and the prefetch/caching problem can be solved in worst case time O(n 3 ) improving over respectively O(n 10 ) [4] and O * (n 18 ) [2]. Implementations are available from the authors home-pages.…”

Dma-based prefetching for i/o-intensive workloads on the cell architecture

“…Alternatively, dynamic prefetching has been proposed that detects applications' reference patterns at runtime, e.g., prediction using probability graphs [21,50], and time series modeling [47]. Prefetch algorithms tailored for parallel I/O systems also have been studied [2,30,31]. Speculative prefetching at the level of whole files or database objects has been proposed by many works [15,20,34,35,38].…”

“…Based on these interactions, a number of works have proposed integrated caching and prefetching schemes [2,11,30,31,32,39,46] that simultaneously identify and handle temporal and spatial I/O access patterns. FlexiCache [33] provides a new flexible interface that allows easy modification of disk cache management decisions using OS-level modules.…”

Strongly competitive algorithms for caching with pipelined prefetching

“…1 Part of this work was done while the author was on a leave at Bell Laboratories, Lucent Technologies, 600 Mountain Ave., Murray Hill, NJ 07974. tween the two was not well understood until the seminal work of Cao et al [6], who proposed to integrate caching with prefetching. They introduced the following execution model.…”