A fast approach to computing exact solutions to the resource-constrained scheduling problem

Abstract

This article presents an algorithm that substantially reduces the computational effort required to obtain the exact solution to the Resource Constrained Scheduling (RCS) problem. The reduction is obtained by (a) using a branch-and-bound search technique, which computes both upper and lower bounds, and (b) using efficient techniques to accurately estimate the possible time-steps at which each operation can be scheduled and using this to prune the search space. Results on several benchmarks with varying resource constraints indicate the clear superiority of the algorithm presented here over traditional approaches using integer linear programming, with speed-ups of several orders of magnitude.

Efficient Resource Constrained Scheduling Using Parallel Structure-Aware Pruning Techniques

“…To effectively avoid unnecessary enumeration of inferior schedules, Narasimhan and Ramanujam [7] presented a B&B approach called BULB using both lower-and upper-bound information to prune the search space. In [24], Hansen and Singh proposed an efficient B&B approach that can reduce the scheduling time considering various resource constraints.…”

“…RCS is an NP-Complete problem [5], [6]. Instead of enumerating all feasible schedules, various approaches [7], [8] are proposed to efficiently prune inferior solutions for RCS. Branch-and-bound (B&B) approaches [7], [9] are promising in existing RCS approaches to prune the search space.…”

“…Instead of enumerating all feasible schedules, various approaches [7], [8] are proposed to efficiently prune inferior solutions for RCS. Branch-and-bound (B&B) approaches [7], [9] are promising in existing RCS approaches to prune the search space. During the state space exploration, the B&B methods update the upper bound information of the optimal schedule dynamically when encountering a better schedule with shorter length.…”

Optimal Superblock Scheduling Using Enumeration

“…Researchers have applied enumeration to instruction scheduling but only in its local form [5,16,19]. A particularly important idea in enumerative approaches to optimal instruction scheduling is using relaxed scheduling as the lower-bound method [16]. Another powerful idea is using history information to prune certain tree nodes if they are dominated by previously visited nodes [19].…”

“…The efficiency is achieved by using a number of pruning techniques that have been proposed in previous work for solving the more fundamental problem of basic-block scheduling [5,16,19]. The contribution of this paper can be viewed as a transformation that decomposes a superblock scheduling problem into a set of basic-block scheduling problems that can be solved optimally using enumeration.…”