https://ira.lib.polyu.edu.hk/bitstream/10397/1269/1/SetupSurvey-2006%28June-19%29.pdf

A survey of scheduling problems with setup times or costs

In this paper, we present a measurement study of the energy consumption characteristics of three widespread mobile networking technologies: 3G, GSM, and WiFi. We find that 3G and GSM incur a high tail energy overhead because of lingering in high power states after completing a transfer. Based on these measurements, we develop a model for the energy consumed by network activity for each technology.Using this model, we develop TailEnder, a protocol that reduces energy consumption of common mobile applications. For applications that can tolerate a small delay such as e-mail, TailEnder schedules transfers so as to minimize the cumulative energy consumed meeting user-specified deadlines. We show that the TailEnder scheduling algorithm is within a factor 2x of the optimal and show that any online algorithm can at best be within a factor 1.62x of the optimal. For applications like web search that can benefit from prefetching, TailEnder aggressively prefetches several times more data and improves user-specified response times while consuming less energy. We evaluate the benefits of TailEnder for three different case study applications - email, news feeds, and web search - based on real user logs and show significant reduction in energy consumption in each case. Experiments conducted on the mobile phone show that TailEnder can download 60% more news feed updates and download search results for more than 50% of web queries, compared to using the default policy.

https://ciir-publications.cs.umass.edu/getpdf.php?id=904

Energy consumption in mobile phones: a measurement study and implications for network applications

In this paper, we present a literature review, classification schemes and a simple meta-analysis for scheduling of batch processors (SBP) research in semiconductor manufacturing (SM). This review is based on a study of journals and web-based documents/articles, which include conference materials, lecture notes in computer science, working papers, etc. There are 98 articles published in various publication outlets between 1986 and October 2004. Based on the literature review carried out and the nature of SBP research observed in SM, we have introduced two classification schemes to systematically organize the published articles. The first classification scheme is based on the problem configurations of SBP research in SM and the second one is based on the solution methodology followed by the researchers. These classification schemes indicate that there is much research scope on SBP research in SM. Furthermore, a simple meta-analysis is carried out to enhance understanding on the development and evolution of SBP research in SM and to identify potential research areas for further research and for improvement. The results show that there is an increasing trend in SBP research in SM. A comprehensive list of references is presented. This study is expected to provide a source of reference for other researchers (or readers), who are interested in SBP research particularly in SM and help stimulate further interest.

A literature review, classification and simple meta-analysis on scheduling of batch processors in semiconductor

Energy-efficient designs have played import roles for hardware and software implementations for a decade. With the advanced technology of VLSI circuit designs, energy-efficiency can be achieved by adopting the dynamic voltage scaling (DVS) technique. In this paper, we survey the studies for energy-efficient scheduling in real-time systems on DVS platforms to cover both theoretical and practical issues.

Energy-Efficient Scheduling for Real-Time Systems on Dynamic Voltage Scaling (DVS) Platforms

In this paper we investigate algorithmic instruments leading to low powerconsumption in computing devices. While previous work on energy-efficient algorithms has mostly focused on single processor environments, in this paper we investigate multi-processor settings. We study the basic problem of scheduling a set of jobs, each specified by a release time, a deadline and a processing volume, on variable speed processors so as to minimize the total energy consumption. We first settle the complexity of speed scaling with unit size jobs. More specifically, we devise a polynomial time algorithm for agreeable deadlines and prove NP-hardness results for arbitrary release dates and deadlines. For the latter setting we also develop a polynomial time algorithm achieving a constant factor approximation guarantee that is independent of the number of processors. Additionally, we study speed scaling of jobs with arbitrary processing requirements and, again, develop constant factor approximation algorithms. We finally transform our offline algorithms into constant competitive online strategies.

http://ce.sharif.ac.ir/~ghodsi/archive/d-papers/SPAA/2007/Speed%20Scaling%20on%20Parallel%20Processors.pdf

Speed scaling on parallel processors

Power Management policies aim at reducing the amount of energy consumed by battery operated systems, while keeping the overall performance high. In this paper we focus on shut-down mechanisms that put a system into a sleep state when it is idle. A very small amount of energy is consumed in this state but, a fixed amount of energy is required when moving the system from the sleep state to the active state. The offline version of this problem consists in scheduling a set of unit execution tasks, with release dates and deadlines, on a single machine in order to minimize the number of idle time periods. We show that this problem can be solved in polynomial time by Dynamic Programming.

http://ce.sharif.ac.ir/~ghodsi/archive/d-papers/ACM%20SODA/2006/Scheduling%20Unit%20Tasks%20to%20Minimize%20the%20Number%20of%20Idle%20Periods.pdf

Scheduling unit tasks to minimize the number of idle periods: a polynomial time algorithm for offline dynamic power management

We study the problems of scheduling jobs, with different release dates and equal processing times, on two types of batching machines. All jobs of the same batch start and are completed simultaneously. On a serial batching machine, the length of a batch equals the sum of the processing times of its jobs and, when a new batch starts, a constant setup time s occurs. On a parallel batching machine, there are at most b jobs per batch and the length of a batch is the largest processing time of its jobs. We show that in both environments, for a large class of so called “ordered” objective functions, the problems are polynomially solvable by dynamic programming. This allows us to derive that the problems where the objective is to minimize the weighted number of late jobs, or the weighted flow time, or the total tardiness, or the maximal tardiness are polynomial. In other words, we show that 1|p-batch,b<n, r

 i, p

 i=p|F and 1|s-batch, r

 i, p

 i=p|F, are polynomial for F∈{∑w

 i

 U

 i,∑w

 i

 C

 i,∑T

 i, T

 max}. The complexity status of these problems was unknown before.

Batching identical jobs

A note on scheduling identical coupled tasks in logarithmic time

Ideal schedules reach both minimum maximum completion time and minimum total completion time of jobs. It is known that there exist computable in polynomial time ideal nonpreemptive two-machine schedules of unit-time operation jobs with equal release dates and arbitrary precedence constraints on identical parallel machines, in flow shops and open shops. In this paper, we study the possibility of extending these results to the case where release dates can be different. We establish the complexity status of P2|prec,r
 j
 ,p
 j
 =1|∑C
 j
 and F2|prec,r
 j
 ,p
 ij
 =1|∑C
 j
 showing that optimal schedules for these problems can also be found in polynomial time and conjecture that all such schedules are ideal indeed. On the other hand, we show that the ideal schedules in open shops do not always exist.

Shortest path to nonpreemptive schedules of unit-time jobs on two identical parallel machines with minimum total completion time

Motivated by the problem of computing trajectories of a set of aircraft in their final descent, we introduce the K king problem, a dramatic simplification of the initial problem in which time and space are discretized. A constraint-based model relying on several specific global constraints is introduced. Computational experiments are reported and show that small instances of this problem can be solved in reasonable time.

Philippe Baptiste

Papers

Scheduling unit tasks to minimize the number of idle periods: a polynomial time algorithm for offline dynamic power management

Batching identical jobs

A note on scheduling identical coupled tasks in logarithmic time

Shortest path to nonpreemptive schedules of unit-time jobs on two identical parallel machines with minimum total completion time

The K King Problem, an Abstract Model for Computing Aircraft Landing Trajectories: On Modeling a Dynamic Hybrid System with Constraints