Race to idle: new algorithms for speed scaling with a sleep state
Susanne Albers,Antonios Antoniadis +1 more
- pp 1266-1285
Reads0
Chats0
TLDR
In this article, the authors considered the energy conservation problem with a variable-speed processor equipped with a sleep state and derived an approximation factor of 4/3 for general convex power functions, and showed that no algorithm that minimizes the energy expended for processing jobs can attain an approximation ratio smaller than 2.Abstract:
We study an energy conservation problem where a variable-speed processor is equipped with a sleep state. Executing jobs at high speeds and then setting the processor asleep is an approach that can lead to further energy savings compared to standard dynamic speed scaling. We consider classical deadline-based scheduling, i.e. each job is specified by a release time, a deadline and a processing volume. For general convex power functions, Irani et al. [12] devised an offline 2-approximation algorithm. Roughly speaking, the algorithm schedules jobs at a critical speed Scrit that yields the smallest energy consumption while jobs are processed. For power functions P(s) = sα + γ, where s is the processor speed, Han et al. [11] gave an (αα + 2)-competitive online algorithm.We investigate the offline setting of speed scaling with a sleep state. First we prove NP-hardness of the optimization problem. Additionally, we develop lower bounds, for general convex power functions: No algorithm that constructs Scrit-schedules, which execute jobs at speeds of at least scrit, can achieve an approximation factor smaller than 2. Furthermore, no algorithm that minimizes the energy expended for processing jobs can attain an approximation ratio smaller than 2.We then present an algorithmic framework for designing good approximation algorithms. For general convex power functions, we derive an approximation factor of 4/3. For power functions P(s) = βsα + γ, we obtain an approximation of 137/117read more
Citations
More filters
Proceedings ArticleDOI
Computational sprinting on a hardware/software testbed
Arun Raghavan,Laurel Emurian,Lei Shao,Marios C. Papaefthymiou,Kevin P. Pipe,Thomas F. Wenisch,Milo M. K. Martin +6 more
TL;DR: This work investigates sprinting using a hardware/software testbed, develops a sprint-aware task-based parallel runtime, finds that maximal-intensity sprinting is not always best, introduces the concept of sprint pacing, and evaluates an adaptive policy for selecting sprint intensity.
Journal ArticleDOI
Greener, Energy-Efficient and Sustainable Networks: State-Of-The-Art and New Trends.
TL;DR: Different paradigms for wireless access networks such as millimetre-wave communications, Long-Term Evolution in unlicensed spectrum, ultra-dense heterogeneous networks, device-to-device communications and massive multiple-input multiple-output communications have been analysed as possible technologies for improvement of wireless networks energy efficiency.
Journal ArticleDOI
Energy Efficient Computing, Clusters, Grids and Clouds: A Taxonomy and Survey
TL;DR: Key findings include: in clusters and grids, use of system level efficiency techniques might increase their energy consumption; in (virtualized) clouds, efficient scheduling and resource allocation can lead to substantially greater economies than consolidation through migration; and in production clouds, performance is affected due to demand fluctuation.
Journal ArticleDOI
On the Interplay between Global DVFS and Scheduling Tasks with Precedence Constraints
TL;DR: An in-depth theoretical study of the more commonly available global DVFS that makes such changes for the entire chip, and a bound on the maximal relative deviation is given.
Book ChapterDOI
Multi Objective Optimization of HPC Kernels for Performance, Power, and Energy
TL;DR: This work empirically examines a variety of metrics, architectures, and code optimization decisions and provides evidence that such tradeoffs exist in practice and helps one explore potential tradeoffs among multiple objectives.
References
More filters
Proceedings ArticleDOI
A scheduling model for reduced CPU energy
TL;DR: This paper proposes a simple model of job scheduling aimed at capturing some key aspects of energy minimization, and gives an off-line algorithm that computes, for any set of jobs, a minimum-energy schedule.
Proceedings ArticleDOI
Optimal power allocation in server farms
TL;DR: The analysis shows that the optimal power allocation is non-obvious and depends on many factors such as the power-to-frequency relationship in the processors, the arrival rate of jobs, the maximum server frequency, the lowest attainable server frequency and the server farm configuration.
Journal ArticleDOI
Speed scaling to manage energy and temperature
TL;DR: The study of speed scaling to manage temperature is initiated and it is shown that the optimal temperature schedule can be computed offline in polynomial-time using the Ellipsoid algorithm and that no deterministic online algorithm can have a better competitive ratio.
Journal ArticleDOI
Algorithmic problems in power management
Sandy Irani,Kirk Pruhs +1 more
TL;DR: This survey places more concentration on lines of research of the authors: managing power using the techniques of speed scaling and power-down which are also currently the dominant techniques in practice.
Journal ArticleDOI
Algorithms for power savings
TL;DR: This paper examines two different mechanisms for saving power in battery-operated embedded systems and gives an off line algorithm which is within a factor of three of the optimal algorithm and an online algorithm with a constant competitive ratio.