Abstract. We study online scheduling to minimize flow time plus energy usage in the dynamic speed scaling model. We devise new speed scaling functions that depend on the number of active jobs, replacing the existing speed scaling functions in the literature that depend on the remaining work of active jobs. The new speed functions are more stable and also more efficient. They can support better job selection strategies to improve the competitive ratios of existing algorithms [5,8], and, more importantly, to remove the requirement of extra speed. These functions further distinguish themselves from others as they can readily be used in the non-clairvoyant model (where the size of a job is only known when the job finishes). As a first step, we study the scheduling of batched jobs (i.e., jobs with the same release time) in the non-clairvoyant model and present the first competitive algorithm for minimizing flow time plus energy (as well as for weighted flow time plus energy); the performance is close to optimal. 1

Abstract. We consider online scheduling algorithms in the dynamic speed scaling model, where a processor can scale its speed between 0 and some maximum speed T. The processor uses energy at rate s α when run at speed s, where α> 1 is a constant. Most modern processors use dynamic speed scaling to manage their energy usage. This leads to the problem of designing execution strategies that are both energy efficient, and yet have almost optimum performance. We consider two problems in this model and give essentially optimum possible algorithms for them. In the first problem, jobs with arbitrary sizes and deadlines arrive online and the goal is to maximize the throughput, i.e. the total size of jobs completed successfully. We give an algorithm that is 4-competitive for throughput and O(1)-competitive for the energy used. This improves upon the 14 throughput competitive algorithm of Chan et al. [10]. Our throughput guarantee is optimal as any online algorithm must be at least 4-competitive even if the energy concern is ignored [7]. In the second problem, we consider optimizing the trade-off between the total flow time incurred and the energy consumed by the jobs. We give a 4-competitive algorithm to minimize total flow time plus energy for unweighted unit size jobs, and a (2 + o(1))α / ln α-competitive algorithm to minimize fractional weighted flow time plus energy. Prior to our work, these guarantees were known only when the processor speed was unbounded (T = ∞) [4]. 1

Energy usage has been an important concern in recent research on online scheduling. In this paper we extend the study of the tradeoff between flow time and energy from the single-processor setting [8, 6] to the multi-processor setting. Our main result is an analysis of a simple non-migratory online algorithm called CRR (classified round robin) on m ≥ 2 processors, showing that its flow time plus energy is within O(1) times of the optimal non-migratory offline algorithm, when the maximum allowable speed is slightly relaxed. This result still holds even if the comparison is made against the optimal migratory offline algorithm (the competitive ratio increases by a factor of 2.5). As a special case, our work also contributes to the traditional online flow-time scheduling. Specifically, for minimizing flow time only, CRR can yield a competitive ratio one or even arbitrarily smaller than one, when using sufficiently faster processors. Prior to our work, similar result is only known for online algorithms that needs migration [21, 23], while the best non-migratory result can achieve an O(1) competitive ratio [14]. The above result stems from an interesting observation that there always exists some optimal migratory schedule S that can be converted (in an offline sense) to a non-migratory schedule S ′ with a moderate increase in flow time plus energy. More importantly, this non-migratory schedule always dispatches jobs in the same way as CRR.

Abstract. Speed scaling is a power management technique that involves dynamically changing the speed of a processor. This gives rise to dualobjective scheduling problems, where the operating system both wants to conserve energy and optimize some Quality of Service (QoS) measure of the resulting schedule. In the most investigated speed scaling problem in the literature, the QoS constraint is deadline feasibility, and the objective is to minimize the energy used. The standard assumption is that the power consumption is the speed to some constant power α. We give the first non-trivial lower bound, namely e α−1 /α, on the competitive ratio for this problem. This comes close to the best upper bound which is about 2e α+1. We analyze a natural class of algorithms called qOA, where at any time, the processor works at q ≥ 1 times the minimum speed required to ensure feasibility assuming no new jobs arrive. For CMOS based processors, and many other types of devices, α = 3, that is, they satisfy the cube-root rule. When α = 3, we show that qOA is 6.7-competitive, improving upon the previous best guarantee of 27 achieved by the algorithm Optimal Available (OA). So when the cube-root rule holds, our results reduce the range for the optimal competitive ratio from [1.2, 27] to [2.4, 6.7]. We also analyze qOA for general α and give almost matching upper and lower bounds. 1

Abstract. In this paper we extend the study of flow-energy scheduling to a model that allows both sleep management and speed scaling. Our main result is a sleep management algorithm called IdleLonger, which works online for a processor with one or multiple levels of sleep states. The design of IdleLonger is interesting; among others, it may force the processor to idle or even sleep even though new jobs have already arrived. IdleLonger works in both clairvoyant and non-clairvoyant settings. We show how to adapt two existing speed scaling algorithms AJC [15] (clairvoyant) and LAPS [9] (non-clairvoyant) to the new model. The adapted algorithms, when coupled with IdleLonger, are shown to be O(1)-competitive clairvoyant and non-clairvoyant algorithms for minimizing flow plus energy on a processor that allows sleep management and speed scaling. The above results are based on the traditional model with no limit on processor speed. If the processor has a maximum speed, the problem becomes more difficult as the processor, once overslept, cannot rely on unlimited extra speed to catch up the delay. Nevertheless, we are able to enhance IdleLonger and AJC so that they remain O(1)-competitive for flow plus energy under the bounded speed model. Non-clairvoyant scheduling in the bounded speed model is left as an open problem. 1

Energy consumption has become an important issue in the study of processor scheduling. Energy reduction can be achieved by allowing a processor to vary the speed dynamically (dynamic speed scaling) [2–4, 7, 10] or to enter a sleep state [1, 5, 8]. In the past, these two mechanisms are often studied separately. It is indeed natural to consider an integrated model in which a

Abstract. The past few years have witnessed different scheduling algorithms for a processor that can manage its energy usage by scaling dynamically its speed. In this paper we attempt to extend such work to the two-processor setting. Specifically, we focus on deadline scheduling and study online algorithms for two processors with an objective of maximizing the throughput, while using the smallest possible energy. The motivation comes from the fact that dual-core processors are getting common nowadays. Our first result is a new analysis of the energy usage of the speed function OA [15,4,8] with respect to the optimal twoprocessor schedule. This immediately implies a trivial two-processor algorithm that is 16-competitive for throughput and O(1)-competitive for energy. A more interesting result is a new online strategy for selecting jobs for the two processors. Together with OA, it improves the competitive ratio for throughput from 16 to 3, while increasing that for energy by a factor of 2. Note that even if the energy usage is not a concern, no algorithm can be better than 2-competitive with respect to throughput. 1

Energy has become a scarce and expensive resource. There is a growing awareness in society that energy saving is a critical issue. This paper surveys algorithmic solutions to reduce energy consumption in computing environments. We focus on the system and device level. More specifically, we study power-down mechanisms as well as dynamic speed scaling techniques in modern microprocessors.

Abstract. We consider the setting of a device that obtains it energy from a battery and some regenerative source such as a solar cell. We consider the speed scaling problem of scheduling a collection of tasks with release times, deadlines, and sizes so as to minimize the energy recharge rate of the regenerative source. This is the first theoretical investigation of speed scaling for devices with a regenerative energy source. We show that the problem can be expressed as a polynomial sized convex program. We show that using the KKT conditions, one can obtain an efficient algorithm to verify the optimality of a schedule. We show that the energy optimal YDS schedule, is 2-approximate with respect to the recharge rate. We show that the online algorithm BKP is O(1)-competitive with respect to recharge rate. 1

potential functions are used to show that a particular online algorithm is locally competitive in an amortized sense. Algorithm analyses using potential functions are sometimes criticized as seeming to be black magic