Scheduling with testing is a recent online problem within the framework of explorable uncertainty motivated by environments where some preliminary action can influence the duration of a task. Jobs have an unknown processing time that can be explored by running a test. Alternatively, jobs can be executed for the duration of a given upper limit. We consider this problem within the setting of multiple identical parallel machines and present competitive deterministic algorithms and lower bounds for the objective of minimizing the makespan of the schedule. In the non-preemptive setting, we present the SBS algorithm whose competitive ratio approaches $3.1016$ if the number of machines becomes large. We compare this result with a simple greedy strategy and a lower bound which approaches $2$. In the case of uniform testing times, we can improve the SBS algorithm to be $3$-competitive. For the preemptive case we provide a $2$-competitive algorithm and a tight lower bound which approaches the same value.
翻译:与测试挂钩是最近一个在线问题,这是在一些初步行动可以影响任务持续时间的环境所驱动的探索性不确定性框架内的近期问题。 工作有未知的处理时间,可以通过测试来探索。 工作有未知的处理时间, 也可以在某一上限的期限内执行工作。 或者, 我们可以在多个相同的平行机器的设置中考虑这一问题, 并提出具有竞争力的确定算法和较低界限, 以尽量减少时间表的扩大。 在非先发制人的环境中, 我们提出SBS算法, 如果机器数量大, 其竞争性比率接近3. 1016美元。 我们用简单的贪婪策略来比较这一结果, 其下限则接近2美元。 在统一的测试时间中, 我们可以将SBS算法改进为3美元竞争。 对于先发制人的情况, 我们提供价值为2美元的竞争性算法和接近相同价值的紧要低的下限。