On the Performance of Greedy Algorithms for Power Consumption Minimization

Benoît, Anne; Renaud-Goud, Paul; Robert, Yves

doi:10.1109/icpp.2011.11

Cited by 5 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bansal and Pruhs (2010) investigated several server scheduling problems, where the standard approach for balancing average QoS (Quality of Service) and fairness is to optimize the l p -norm. Benoit et al (2011) considered the objective of minimizing the total power consumed by the processors to execute the job, which is equivalent to the minimization of l 3 -norm.…”

Section: Literature Review Of Hierarchical Schedulingmentioning

confidence: 99%

Semi-Online Hierarchical Scheduling on Two Machines for lp-Norm Load Balancing

Yuan

2019

Asia Pac. J. Oper. Res.

View full text Add to dashboard Cite

This paper investigates semi-online hierarchical scheduling problems on two identical machines, with the purpose of minimizing the [Formula: see text]-norm of the machines’ loads. We consider two semi-online versions with knowing the total processing time [Formula: see text] of all jobs, or knowing the total processing time [Formula: see text] of the jobs of hierarchy [Formula: see text] for [Formula: see text] in advance. For the two semi-online versions, the best possible online algorithms are designed with competitive ratios of [Formula: see text] and [Formula: see text], respectively, where [Formula: see text] is the maximum value of the function [Formula: see text] in [Formula: see text]. When [Formula: see text], our results cover the known results for minimizing the makespan.

show abstract

Section: Literature Review Of Hierarchical Schedulingmentioning

confidence: 99%

Semi-Online Hierarchical Scheduling on Two Machines for lp-Norm Load Balancing

Yuan

2019

Asia Pac. J. Oper. Res.

View full text Add to dashboard Cite

show abstract

“…[1] showed that the problem of scheduling independent tasks can be approximated by a factor (1 + ε): they exhibit a polynomial time approximation scheme (PTAS). [9] studied the performance of greedy algorithms for the problem of scheduling independent tasks, with the objective of minimizing the energy consumption, and proposed some approximation algorithms.…”

Section: Introductionmentioning

confidence: 99%

“…One may try to minimize the ℓ k norm, i.e., the quantity ( p q=1 ( i∈load(q) a i ) k ) 1/k , with p processors, where i ∈ load(q) means that task T i is assigned to processor q, and a i is the weight of task T i [1]. Minimizing the power consumption then amounts to minimize the ℓ 3 norm [9], and the problem of makespan minimization is equivalent to minimizing the ℓ ∞ norm, i.e., minimize max 1≤q≤p i∈load(q) a i [13,5]. These problems are typical load balancing problems, in which the load (computation requirement of the tasks) must be balanced between processors, according to various criteria.…”

Section: Introductionmentioning

confidence: 99%

Approximation Algorithms for Energy, Reliability, and Makespan Optimization Problems

Aupy

Benoît

2016

Parallel Process. Lett.

Self Cite

View full text Add to dashboard Cite

In this paper, we consider the problem of scheduling an application on a parallel computational platform. The application is a particular task graph, either a linear chain of tasks, or a set of independent tasks. The platform is made of identical processors, whose speed can be dynamically modified. It is also subject to failures: if a processor is slowed down to decrease the energy consumption, it has a higher chance to fail. Therefore, the scheduling problem requires to re-execute or replicate tasks (i.e., execute twice a same task, either on the same processor, or on two distinct processors), in order to increase the reliability. It is a tri-criteria problem: the goal is to minimize the energy consumption, while enforcing a bound on the total execution time (the makespan), and a constraint on the reliability of each task.Our main contribution is to propose approximation algorithms for these particular classes of task graphs. For linear chains, we design a fully polynomial time approximation scheme. However, we show that there exists no constant factor approximation algorithm for independent tasks, unless P=NP, and we are able in this case to propose an approximation algorithm with a relaxation on the makespan constraint.

show abstract

“…e.g. [17]), which results in the relation between power cost and transmission speed given as P ∼ f 3 ( [4,5]). Such a model of power consumption was recently studied in the context of splittable Manhattan-path routing by Benoit et al [4].…”

Section: Introductionmentioning

confidence: 99%

Splittable Single Source-Sink Routing on CMP Grids: A Sublinear Number of Paths Suffice

Kosowski

Uznański

2013

Euro-Par 2013 Parallel Processing

View full text Add to dashboard Cite

In single chip multiprocessors (CMP) with grid topologies, a significant part of power consumption is attributed to communications between the cores of the grid. We investigate the problem of routing communications between CMP cores using shortest paths, in a model in which the power cost associated with activating a communication link at a transmission speed of f bytes/second is proportional to f α , for some constant exponent α > 2.Our main result is a trade-off showing how the power required for communication in CMP grids depends on the ability to split communication requests between a given pair of node, routing each such request along multiple paths. For a pair of cores in a m × n grid, the number of available communication paths between them grows exponentially with n, m. By contrast, we show that optimal power consumption (up to constant factors) can be achieved by splitting each communication request into k paths, starting from a threshold value of k = Θ(n 1/(α−1) ). This threshold is much smaller than n for typical values of α ≈ 3, and may be considered practically feasible for use in routing schemes on the grid. More generally, we provide efficient algorithms for routing multiple k-splittable communication requests between two cores in the grid, providing solutions within a constant approximation of the optimum cost. We support our results with algorithm simulations, showing that for practical instances, our approach using k-splittable requests leads to a power cost close to that of the optimal solution with arbitrarily splittable requests, starting from the stated threshold value of k.

show abstract

On the Performance of Greedy Algorithms for Power Consumption Minimization

Cited by 5 publications

References 19 publications

Semi-Online Hierarchical Scheduling on Two Machines for lp-Norm Load Balancing

Semi-Online Hierarchical Scheduling on Two Machines for lp-Norm Load Balancing

Approximation Algorithms for Energy, Reliability, and Makespan Optimization Problems

Splittable Single Source-Sink Routing on CMP Grids: A Sublinear Number of Paths Suffice

Contact Info

Product

Resources

About