Next Article in Journal
Lifetime Prediction of a Polymer Electrolyte Membrane Fuel Cell under Automotive Load Cycling Using a Physically-Based Catalyst Degradation Model
Previous Article in Journal
A Fractional Order Power System Stabilizer Applied on a Small-Scale Generation System
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Productive Efficiency of Energy-Aware Data Centers

Department of Computer Languages and Systems, University of Seville, 41012 Sevilla, Spain
Department of Applied Economy I, University of Seville, 41018 Sevilla, Spain
Author to whom correspondence should be addressed.
Energies 2018, 11(8), 2053;
Received: 17 July 2018 / Revised: 27 July 2018 / Accepted: 1 August 2018 / Published: 8 August 2018
(This article belongs to the Section A: Sustainable Energy)


Information technologies must be made aware of the sustainability of cost reduction. Data centers may reach energy consumption levels comparable to many industrial facilities and small-sized towns. Therefore, innovative and transparent energy policies should be applied to improve energy consumption and deliver the best performance. This paper compares, analyzes and evaluates various energy efficiency policies, which shut down underutilized machines, on an extensive set of data-center environments. Data envelopment analysis (DEA) is then conducted for the detection of the best energy efficiency policy and data-center characterization for each case. This analysis evaluates energy consumption and performance indicators for natural DEA and constant returns to scale (CRS). We identify the best energy policies and scheduling strategies for high and low data-center demands and for medium-sized and large data-centers; moreover, this work enables data-center managers to detect inefficiencies and to implement further corrective actions.

1. Introduction

Data centers, which constitute the computational muscle for cloud computing, can be compared in energy consumption to many industrial facilities and towns. The latest trends show that these infrastructures represent approximately 2% of global energy consumption [1], with a 5% annual growth rate [2].
The data envelopment analysis mathematical model enables the management organizational divisions to measure the performance of an organization by providing the relative efficiency of each organizational unit. This relative efficiency measurement can be applied to a set of decision-making units, also known as DMUs, or for productive efficiency. The productive efficiency, also called technical efficiency, involves a collection of inputs (the resources needed for the production) and outputs (the production achieved). To this end, DEA constructs an “efficiency frontier” which places the relative performance of all units so these can be contrasted. This method is notably well-suited for the examination of the behavior of complex relations, even unknown, between numerous inputs and outputs, where the decisions made are affected by a level of uncertainty [3]. Moreover, DEA has been used both in private [4,5] and in public contexts [6,7,8,9].
Many initiatives have emerged looking for the decrease of the consumption of energy and the CO 2 trace of data-centers, especially those of a medium and large size. These facilities are composed of thousands and even tens of thousands of machines.
A substantial part of these initiatives focuses on the improvement of the Power Usage Effectiveness (PUE), that is the amount of energy consumed in non-computational tasks, such as power supply, cooling and networking components. This accounts for more than half of the energy consumption of an Internet data-center (IDC).
Several strategies are proposed to significantly improve energy efficiency in large-scale clusters [10]: cooling and temperature management [11,12]; power proportionality for CPU and memory hardware components [13,14]; fewer energy-hungry and non-mechanical hard disks [15]; and new proposals for energy distribution [16].
On the other hand, almost 50% of energy is consumed by computational servers to satisfy the incoming workload. The job arrival is not stable over time, but usually presents correlative low and high periods, such as those present in day/night and weekday/weekend workload patterns.
Such scenarios present a huge opportunity for the improvement of energy efficiency through proper scheduling and through the application of low-energy consumption modes to servers, since keeping servers in an idle state is extremely energy-inefficient. Many energy-aware schedulers, which aim to raise server usage, have been proposed in order to free up the maximum amount of machines so that they may put into hibernation [17,18,19]. In addition to these schedulers, several energy-conservation strategies may be applied in virtualized environments, such as the consolidation and migration of virtual machines [20,21].
Other strategies focus on the reduction of energy consumption in specific scenarios, such as those of distributed file systems [22,23].
The most aggressive approach involves the shut-down of underutilized servers in order to minimize energy consumption. Several shut-down policies have been proposed for grid computing environments in [24]. This strategy is yet to be widely implemented in working data-centers since a natural reticence to worsening QoS is usually present in data-center operators [25].
The innovation of the research presented in this paper involves the utilization of data envelopment analysis (DEA) as a mathematical technique to compare the efficiency regarding the consumption of energy and the performance of various workload scenarios, scheduling models and energy efficiency policies. This efficiency analysis enables data-center operators to make appropriate decisions about the number of machines, the scheduling solution and the shut-down strategy that must be applied so that data-centers run optimally. The final goal is the maximization of the productive efficiency, which is computed as the amount of energy consumed to serve a workload with a determined performance.
The major contributions of this paper can be summarized as follows:
  • Extensive empirical experimentation and analysis of various cloud-computing scenarios with a trustworthy and detailed simulation tool.
  • Impact analysis in terms of the energy consumption and performance of several energy efficiency policies, which shut-down idle machines by means of data envelopment analysis.
  • DEA-conducted analysis of the performance impact and energy consumption of a set of scheduling models for large-scale data-centers.
  • Empirical determination and proposal of corrective actions to achieve optimal efficiency.
The work is organized as follows. In Section 2, the authors introduce the current literature for the utilization of DEA presented for various areas, as well as the DEA model employed in this work. In Section 3, we briefly explain the set of energy efficiency policies that shut down idle servers. The scheduling models considered are explained in Section 4. In Section 5, the tool used for the simulation, the experimental environment, the energy model and DEA inputs/outputs are presented. Natural constant returns to scale (CRS) DEA results are described and analyzed in Section 6. Finally, we summarize this paper and present conclusions in Section 7.

2. Data Envelopment Analysis Model

Data envelopment Analysis (DEA) is a method that analyzes the connections between the outputs and inputs required in a production process in order to establish the efficiency frontiers [26]. This non-parametric technique was first described for the determination of the efficiency of DMUs by [27] and was formally defined by [28]. DEA has been proposed to measure the efficiency in various areas of operations research and management science [29,30,31,32]. Moreover, it has been applied to measure the environmental performance by other authors [33,34,35,36,37,38,39,40], who describe the gains of this method in the field of environmental management, which is a matter of undoubted relevance for the valuation of the sustainable development ability and pathway [41]. A critical feature of DEA for environmental analysis is the inclusion of desirable and undesirable outputs along with its own production variables, which cannot be isolated in an environmental analysis model of these features [42]. In this way, ref. [36] have refined a non-radial and radial model of DEA for environmental measurements. This approach separates the outputs into desirable and undesirable and presents two concepts: natural and managerial disposability. In this work, we employ the DEA radial approach for environmental assessments proposed by [37]. It should be borne in mind that a main feature of this approach is the utilization of DEA-RAM (range-adjusted measure), first proposed by [43] to treat in a unified manner the analysis of managerial and natural disposability.

2.1. Natural Disposability

Natural disposability refers to a DMU that improves its efficiency by decreasing its inputs in order to decrease its undesirable outputs, as well as to increase the desirable outputs.
In Model (1), each j-th DMU j = 1 , , n , considers inputs X j = ( x 1 j , , x m j ) T for the production of desirable outputs G j = ( g 1 j , , g s j ) T and undesirable outputs B j = ( b 1 j , , b h j ) T . Furthermore, d i x , i = 1 , , m , d r g , r = 1 , , s and d f b , f = 1 , , h are all slack variables which are related to inputs, desirable and undesirable outputs, respectively. λ = ( λ 1 , , λ n ) T are structural or intensity variables, which are unknown and are used for the connection of the input and output vectors by means of a convex combination. R is the range resolute through the lower and upper limits of inputs, desirable outputs and undesirable outputs, denoted by:
R i x = ( m + s + h ) 1 ( max { x i j / j = 1 , , n } min { x i j / j = 1 , , n } )
R r g = ( m + s + h ) 1 ( max { g r j / j = 1 , , n } min { g r j / j = 1 , , n } ) a n d
R f b = ( m + s + h ) 1 ( max { b f j / j = 1 , , n } min { b f j / j = 1 , , n } )
The natural efficiency of the k-th policy is computed by the following CRS and radial VRS model (see [37] for a better understanding):
max ξ + ϵ ( i = 1 m R i x d i x + r = 1 s R r g d i g + f = 1 h R f b d i b )
s . t . j = 1 n x i j λ j + d i x = x i k , i = 1 , , m , j = 1 n g r j λ j d r g ξ g r k = g r k , r = 1 , , s , j = 1 n b f j λ j + d f b + ξ b f k = b f k , f = 1 , , h , d i x 0 , i = 1 , , m , d r g 0 , r = 1 , , s , d f b 0 , f = 1 , , h , ξ Unrestricted
where the unrestricted parameter ξ denotes an unknown inefficiency rate expressing the gap between the efficiency frontier and an empirical group of undesirable and desirable outputs. The parameter ϵ takes the value of 0.0001 in this work to minimize the influence of slack variables. If the restriction j = 1 n λ j = 1 is added to Model (1), then the obtained model is a VRS (Model (1 * )).
The first restriction in equation systems ((1), (1 * )) explores the values of λ j to create a composite unit, considering inputs such as: j = 1 n x i j λ j = d i x + x i k , i = 1 , , m . The values of the inputs can be decreased when the positive slack variables d i x are present. This may unquestionably vary the given rates, which implies that the system presents some inefficiencies.
In the same way, the second restriction, j = 1 n g r j λ j = d r g + ξ g r k + g r k , r = 1 , , s , indicates that the desirable outputs can be maintained or increased by making an increase of the slack variable d r g and a radial expansion ξ g r k .
The third restriction, j = 1 n b f j λ j = d f b ξ b f k + b f k , f = 1 , , h , shows the decrease of the inputs, and then, we could reduce the undesirable outputs both in their slack variables and radially.
The objective function considers that two origins of inefficiency may be established. A k-policy can be considered efficient when the following two conditions are met: (a) ξ = 0 ; (b) d i x = 0 , d r g = 0 , d f b = 0 . In this case, the k-policy belongs to the efficiency frontier, since it fulfills the constraints present in equation systems ((1), (1 * )), and consequently, the objective function takes a value of zero. Otherwise, the value of the objective function for non-efficient policies is greater than zero, due to possible displacements in the slack variables and radial movements.
The natural efficiency is then computed by:
θ * = 1 ξ * + ϵ ( i = 1 m R i x d i x * + r = 1 s R r g d i g * + f = 1 h R f b d i b * )
The value of this unified efficiency measure ranges between zero and one. If the k-policy is efficient, then the objective function of equation systems ((1), (1 * )) is zero, and hence, the efficiency score equals θ * = 1 . Slack variables resulting in the optimality of the models represented in equation systems ((1), (1 * )) show the level of inefficiency.

2.2. Managerial Disposability

The managerial efficiency of the k-th policy is evaluated by the following CRS and VRS radial model [37]:
s . t . j = 1 n x i j λ j d i x = x i k , i = 1 , , m , j = 1 n g r j λ j d r g ξ g r k = g r k , r = 1 , , s , j = 1 n b f j λ j + d f b + ξ b f k = b f k , f = 1 , , h , d i x 0 , i = 1 , , m , d r g 0 , r = 1 , , s , d f b 0 , f = 1 , , h , ξ Unrestricted
Similarly, if the restriction j = 1 n λ j = 1 is added to Model (2), then the obtained model is a VRS (Model (2 * )). In this model (2), increasing the inputs is allowed since new technologies that emit less CO 2 emissions to the atmosphere can be used.
By using the VRS models, we can obtain the returns to scale (RTS) and damage to scale (DTS) (see [37] for a better understanding). It is clear that for the natural efficiency, the returns to scale have to be increasing, and for managerial efficiency, the damages to scale have to be decreasing. Otherwise, the technical units are not working well and should correct the imbalances, using the information of the efficient units to which they have to be similar (peers).

3. Energy Policies for Data Centers at a Glance

The following set of energy efficiency policies for shutting down underutilized machines have been developed in this work as an evolution of those presented in [24], which have been adapted to the more complex reality of the cloud-computing paradigm:
  • Never: prevents any shut-down process.
  • Always: shuts down every server running in an idle state.
  • Load: shuts down machines when data-center load pressure fails to reach a given threshold.
  • Margin: assures that a determined number of machines are turned on and available before shutting down any machine.
  • Random: shuts down machines randomly by means of a Bernoulli distribution with parameter 0 . 5 .
  • Exponential: shuts down machines when the probability of one incoming task negatively impacting on the data-center performance is lower than a given threshold. This probability is computed by means of the exponential distribution.
  • Gamma: shuts down machines when the probability of incoming tasks oversubscribing to the available resources in a particular time period is lower than a given threshold; this probability is computed by means of the Gamma distribution.

4. Scheduling Models for Data Centers at a Glance

Cluster schedulers constitute a core part of cloud computing systems, since they are responsible for optimal task assignation to computing nodes. Several degrees of parallelism have been added to overcome the limitations present in central monolithic scheduling approaches when complex and heterogeneous systems with a high number of incoming jobs are considered. The following scheduling models are studied in this work:
  • Monolithic: A centralized and single scheduler is responsible for scheduling all tasks in the workload in this model [44]. This scheduling approach may be the perfect choice when real-time responses are not required [45,46], since the omniscient algorithm performs high-quality task assignations by considering all restrictions and features of the data-center [47,48,49,50] at the cost of longer latency [46]. The scheduling process of a monolithic scheduler, such as that given by Google Borg [51], is illustrated in Figure 1.
  • Two-level: This model achieves a higher level of parallelism by splitting the resource allocation and the task placement: a central manager blocks the whole cluster every time a scheduler makes a decision to offer computing resources to schedulers; and a set of parallel application-level schedulers performs the scheduling logic against the resources offered. This strategy enables the development of sub-optimal scheduling logic for each application, since the state of the data-center is not shared with the central manager, nor with the application schedulers. The workflow of the Two-level schedulers [53,54] is represented in Figure 2.
  • Shared-state schedulers: On the other hand, in shared-state schedulers, such as Omega [55], the state of the data-center is available to all the schedulers. The central manager coordinates all the simultaneous parallel schedulers, which perform the scheduling logic against an out-of-date copy of the state of the data-center. The scheduling decisions are then committed to the central manager, which strives to apply these decisions. The utilization of stale views of the cluster by the schedulers can result in conflicts, since the chosen resources may not longer be available. In such a scenario, the local view of the state of the data-center stored in the scheduler is refreshed before the repetition of the scheduling process. The workflow of the shared-state scheduling model is represented in Figure 3.

5. Methodology

In these next sections, the experimental environment designed for the implementation of the natural CRS DEA analysis is presented. The workflow followed in this work is shown in Figure 4.

5.1. Simulation Tool

The SCORE simulator [52] is employed in this work, since simulation is the best alternative in scenarios where the implementation of the proposed strategies on real large-scale data-centers remains unfeasible. This simulator provides us with the tools for the development and application of the energy policies described in Section 3 and the scheduling models presented in Section 4 on realistic large-scale cloud computing systems.

5.2. Environment and DMU Definition

Following the trends presented in [56,57], two utilization environments have been simulated in this paper for seven days of operation:
  • the low-utilization scenario, which represents highly over-provisioned infrastructures and achieves an average utilization of approximately 30%.
  • the high-utilization scenario, which represents facilities of a more efficient nature that use approximately 65% of available resources on average.
These scenarios are applied to three data-center sizes: (a) Small: composed of 1000 computing servers; (b) Medium: composed of 5000 computing servers; and (c) Large: composed of 10,000 computing servers. Each server is equipped with four CPU cores and 8 GB of RAM.
Decision-making units (DMUs) are defined by the following elements: (a) an energy efficiency policy; (b) a scheduling model; and (c) a workload scenario.

5.3. Energy Model

The following states are presented for each resource in the energy model applied in this work: (a) Idle: when the machine is not executing tasks; and (b) Busy: otherwise.
Let t i d l e i represent the time the i-th resource is idle, and let t b u s y i denote the time during which the machine is computing tasks. In the same way, P i d l e i and P b u s y i represent the power required for the machines to run in these states, respectively.
The time a machine spends on executing a job may be defined as follows:
t b u s y i j = max t T a s k s i C t
where T a s k s i j represents the tasks of the j-th job assigned to M i and C t denotes the completion time of the t-th task of the j-th job.
In the same way, the total time a machine is executing tasks and the total time it is in an idle state may be defined as follows:
t b u s y i = j = 1 j t b u s y i j
t i d l e i = t t o t a l i t b u s y i
where t t o t a l i represents the total operation time. Therefore, we can express the energy consumption as follows:
i = 1 m ( P b u s y i t b u s y i + P i d l e i t i d l e i )
The considered power states, transitions and values for the energetic model are shown in Figure 5.

5.4. DEA Inputs and Outputs

The inputs and outputs considered in DEA analysis and representative experimentation values are shown in Table 1 and Table 2, respectively. One hundred and eight DMUs were analyzed, which were the result of the combination of all energy policies, scheduling models, data-center sizes and workload types described in Section 3, Section 4 and Section 5.2, respectively. However, for clarity, a subset of the most interesting eighteen DMUs i shown in this paper. Each environment presents the following inputs and outputs:
  • Inputs: Two inputs are considered in this work: (a) the number of machines in the data-center (D.C.), as shown in Section 5.2; and (b) the number of shut-down operations performed. These inputs may be reduced or kept equal.
  • Outputs: One desirable output and two undesirable outputs are considered in this paper: (a) the time used to perform tasks’ operations. The longer the time, the less idle the data-center. This good input can be maximized or kept equal; (b) the energy consumption of the data-center. The lower the energy consumption, the more efficient the data-center. This bad input may be reduced or kept equal; and (c) the average time jobs spend in a queue until they are scheduled. The shorter the time, the more performant the system is. This bad input may be reduced or kept equal.

6. Natural CRS DEA Results

The whole dataset included as an Appendix is analyzed by means of natural CRS and VRS DEA. However, only the most relevant natural CRS DEA results for the most representative DMUs, which are presented in Table 2, are described in this section.
An efficiency analysis depending on the data-center size and on the energy policy is shown in Table 3 and Table 4. The following conclusions can be drawn:
  • The best efficiency levels are achieved for small data-centers. The data-center size input is predominant in this group of DMUs, since no major differences between energy policies, scheduling frameworks and workload scenarios are present ( σ = 0.01, x ¯ = 0.99).
  • Mid-size data-centers should use the margin energy policy and monolithic or Omega schedulers and should avoid all other energy policies and the Mesos scheduler. Moreover, high workload scenarios are also more efficient than low workload scenarios. In addition, the following DMUs achieve a good level of efficiency, but they do not belong to the efficiency frontier: (a) the DMU combining the Gamma energy policy and the monolithic or Omega schedulers; (b) the DMU combining the exponential energy policy and the Omega scheduler.
  • No DMU is efficient in large-scale data-centers. However, the following DMUs present good levels of efficiency: (a) the DMUs combining the Gamma, exponential or margin energy policy with the high workload scenario and the monolithic scheduler; and (b) the DMUs combining the Gamma or margin energy policy with the high workload scenario and the Omega scheduler.
  • In high-loaded scenarios, the monolithic scheduler presents the lowest deviation regardless of the data-center size ( σ = 0.32).
We can determine that it is always inefficient to operate in a low utilization scenario in medium-sized and large data-centers. Moreover, both the margin and the probabilistic energy policies (Gamma and exponential) perform more efficiently than the rest of the energy policies, as shown in Figure 6. The monolithic scheduler seems to achieve good results even for large-scale data-centers, while the two-level scheduling approach has a negative impact on data-center performance. However, the trends show that the performance of the monolithic scheduling approach suffers from degradation on larger data-centers and higher workload pressure, and hence, lower efficiency levels are to be expected if larger sizes and higher utilization scenarios are to be considered.
The actions proposed for the improvement of efficiency of the most relevant DMUs are shown in Table 5.

6.1. Proposed Corrections for a Sample DMU

DMU #104 is selected to illustrate how corrective actions are proposed by DEA in order to achieve efficiency. This DMU is defined by the combination of the random energy efficiency policy, the Omega scheduling model and a low utilization workload scenario.
DMU #104 presents a natural efficiency of 0.1697. This means it is far from being efficient. The following corrective actions are suggested for it to belong to the efficiency frontier, as shown in Table 6:
  • The time the data-center spends on task computation must be increased by 38.28 h (+83%).
  • Energy consumption must be reduced by 193.88 MWh (−83%).
  • The average time jobs wait in a queue must be reduced by 3.23 s (−83%).
  • The number of servers must be reduced by 9190 (−92%).
  • Shut-down operations must be reduced by 9680 (−24%).
In addition to these corrective actions, the peers this DMU should emulate are #13, #34 and #18. This means that the workload must be increased, and better energy efficiency policies, such as margin and always, must be used. The full dataset containing all the DMUs and DEA analysis and corrections can be found as Supplementary Material in the Appendix.
Some of the proposed changes involve the switching of the scheduling framework, which is hardly achievable with the current resource manager systems. To implement these corrections, a resource managing system able to dynamically change the scheduling framework during runtime would be necessary. Such a system is an interesting improvement to the current state of the art that the DEA analysis leads us to develop.

7. Conclusions and Policy Implications

In this work, we have confirmed the hypothesis that DEA constitutes a powerful tool for the analysis of technical efficiency in cloud-computing scenarios where large-scale data-centers provide the computational core.
Data envelopment analysis provides cloud-computing operators with the means for the identification of which data-center configuration better suits their requirements, both in terms of performance and energy efficiency.
This methodology allows us to analyze several energy efficiency policies that shut down idle servers, so that their behavior and differences can be compared in various data-center environments. It has been proven that policies based on a security margin and those that use statistical tools to predict the future workload, such as exponential and Gamma, deliver better results than policies based on data-center workload pressure and random strategies.
In addition, it has been empirically shown that even under medium and high workload pressure, in data-centers composed of up to 10,000 machines, monolithic schedulers perform better than other scheduling models, such as the two-level and shared-state approaches.
Finally, cloud-computing infrastructure managers are provided with empirical knowledge of which data-centers are not being used optimally, and hence, they can make decisions regarding the shut-down of machines in order to achieve higher utilization levels of the cloud-computing system as a whole.
As future work related to the limitations of the presented work, we may include:
  • The addition of different kind of workload patterns, as well as real workload traces.
  • The analysis of other scheduling models, such as distributed and hybrid models.
  • The development of a new-generation resource-managing system that could dynamically apply the optimal scheduling framework depending on the environment and workload.
  • The analysis of simulation data with other DEA approaches, such as Bayesian and probabilistic models, which could minimize the impact of the noise in current DEA models.

Supplementary Materials

Supplementary Materials are available online at

Author Contributions

D.F.-C. and A.F.-M. conceived of and designed the experiments. D.F.-C. performed the experiments. D.F.-C., A.F.-M. and V.F. analyzed the data. D.F.-C. and F.V. contributed reagents/materials/analysis tools. D.F.-C., A.F.-M. and F.V. wrote the paper.


This research was funded by VI Plan Propio de Investigación y Transferencia—University of Seville 2018 grant number 2018/00000520.


The research is supported by the VI Plan Propio de Investigación y Transferencia (VIPPI), University of Seville.

Conflicts of Interest

The authors declare no conflict of interest.


  1. Koomey, J. Growth in Data Center Electricity Use 2005 to 2010; Analytical Press: Piedmont, CA, USA, 1 August 2011. [Google Scholar]
  2. Van Heddeghem, W.; Lambert, S.; Lannoo, B.; Colle, D.; Pickavet, M.; Demeester, P. Trends in worldwide ICT electricity consumption from 2007 to 2012. Comput. Commun. 2014, 50, 64–76. [Google Scholar] [CrossRef][Green Version]
  3. Gómez-López, M.T.; Gasca, R.M.; Pérez-Álvarez, J.M. Decision-Making Support for the Correctness of Input Data at Runtime in Business Processes. Int. J. Cooper. Inf. Syst. 2014, 23. [Google Scholar] [CrossRef]
  4. Amirteimoori, A.; Emrouznejad, A. Optimal input/output reduction in production processes. Decis. Support Syst. 2012, 52, 742–747. [Google Scholar] [CrossRef]
  5. Chiang, K.; Hwang, S.N. Efficiency measurement for network systems IT impact on firm performance. Decis. Support Syst. 2010, 48, 437–446. [Google Scholar]
  6. Chang, Y.T.; Zhang, N.; Danao, D.; Zhang, N. Environmental efficiency analysis of transportation system in China: A non-radial DEA approach. Energy Policy 2013, 58, 277–283. [Google Scholar] [CrossRef]
  7. Arcos-Vargas, A.; Núñez-Hernández, F.; Villa-Caro, G. A DEA analysis of electricity distribution in Spain: An industrial policy recommendation. Energy Policy 2017, 102, 583–592. [Google Scholar] [CrossRef]
  8. Gonzalez-Rodriguez, M.; Velasco-Morente, F.; González-Abril, L. La eficiencia del sistema de protección social español en la reducción de la pobreza. Papeles de Población 2010, 16, 123–154. [Google Scholar]
  9. Afonso, A.; Schuknecht, L.; Tanzi, V. Public sector efficiency: Evidence for new EU member states and emerging markets. Appl. Econ. 2010, 42, 2147–2164. [Google Scholar] [CrossRef]
  10. Jakóbik, A.; Grzonka, D.; Kolodziej, J.; Chis, A.E.; González-Vélez, H. Energy Efficient Scheduling Methods for Computational Grids and Clouds. J. Telecommun. Inf. Technol. 2017, 1, 56–64. [Google Scholar]
  11. Sharma, R.K.; Bash, C.E.; Patel, C.D.; Friedrich, R.J.; Chase, J.S. Balance of power: Dynamic thermal management for internet data centers. IEEE Internet Comput. 2005, 9, 42–49. [Google Scholar] [CrossRef]
  12. El-Sayed, N.; Stefanovici, I.A.; Amvrosiadis, G.; Hwang, A.A.; Schroeder, B. Temperature management in data-centers: Why some (might) like it hot. In Proceedings of the 12th ACM SIGMETRICS/PERFORMANCE Joint International Conference on Measurement and Modeling of Computer Systems, London, UK, 11–15 June 2012; pp. 163–174. [Google Scholar]
  13. Miyoshi, A.; Lefurgy, C.; Van Hensbergen, E.; Rajamony, R.; Rajkumar, R. Critical power slope: Understanding the runtime effects of frequency scaling. In Proceedings of the 16th International Conference on Supercomputing, New York, NY, USA, 22–26 June 2016; pp. 35–44. [Google Scholar]
  14. Fan, X.; Weber, W.D.; Barroso, L.A. Power provisioning for a warehouse-sized computer. In Proceedings of the 34th Annual International Symposium on Computer Architecture, San Diego, CA, USA, 9–13 June 2007; pp. 13–23. [Google Scholar]
  15. Andersen, D.G.; Swanson, S. Rethinking flash in the data-center. IEEE Micro 2010, 30, 52–54. [Google Scholar] [CrossRef]
  16. Femal, M.E.; Freeh, V.W. Boosting data-center performance through non-uniform power allocation. In Proceedings of the Second International Conference on Autonomic Computing (ICAC’05), Seattle, WA, USA, 13–16 June 2005. [Google Scholar] [CrossRef]
  17. Jakóbik, A.; Grzonka, D.; Kołodziej, J. Security supportive energy aware scheduling and scaling for cloud environments. In Proceedings of the 31st European Conference on Modelling and Simulation (ECMS 2017), Budapest, Hungary, 23–26 May 2017; pp. 583–590. [Google Scholar]
  18. Juarez, F.; Ejarque, J.; Badia, R.M. Dynamic energy-aware scheduling for parallel task-based application in cloud computing. Future Gener. Comput. Syst. 2018, 78, 257–271. [Google Scholar] [CrossRef][Green Version]
  19. Lee, Y.C.; Zomaya, A.Y. Energy efficient utilization of resources in cloud computing systems. J. Supercomput. 2012, 60, 268–280. [Google Scholar] [CrossRef]
  20. Sohrabi, S.; Tang, A.; Moser, I.; Aleti, A. Adaptive virtual machine migration mechanism for energy efficiency. In Proceedings of the 5th International Workshop on Green and Sustainable Software, Austin, TX, USA, 14–22 May 2016; pp. 8–14. [Google Scholar]
  21. Beloglazov, A.; Buyya, R. Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in cloud data-centers. Concurr. Comp.-Pract. E 2012, 24, 1397–1420. [Google Scholar] [CrossRef]
  22. Kaushik, R.T.; Bhandarkar, M. Greenhdfs: Towards an energy-conserving, storage-efficient, hybrid hadoop compute cluster. In Proceedings of the 2010 International Conference on Power Aware Computing and Systems, Vancouver, BC, Canada, 3–6 October 2010; pp. 1–9. [Google Scholar]
  23. Luo, X.; Wang, Y.; Zhang, Z.; Wang, H. Superset: A non-uniform replica placement strategy towards high-performance and cost-effective distributed storage service. In Proceedings of the 2013 International Conference on Advanced Cloud and Big Data, Nanjing, China, 13–15 December 2013. [Google Scholar] [CrossRef]
  24. Fernández-Montes, A.; Gonzalez-Abril, L.; Ortega, J.A.; Lefèvre, L. Smart scheduling for saving energy in grid computing. Expert Syst. Appl. 2012, 39, 9443–9450. [Google Scholar] [CrossRef]
  25. Fernández-Montes, A.; Fernández-Cerero, D.; González-Abril, L.; Álvarez-García, J.A.; Ortega, J.A. Energy wasting at internet data-centers due to fear. Pattern Recogn. Lett. 2015, 67, 59–65. [Google Scholar] [CrossRef]
  26. Farrell, M.J. The measurement of productive efficiency. J. R. Stat. Soc. Ser. A (Gen.) 1957, 120, 253–290. [Google Scholar] [CrossRef]
  27. Charnes, A.; Cooper, W.W.; Rhodes, E. Measuring the efficiency of decision making units. Eur. J. Oper. Res. 1978, 2, 429–444. [Google Scholar] [CrossRef]
  28. Banker, R.D.; Charnes, A.; Cooper, W.W. Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis. Manag. Sci. 1984, 30, 1078–1092. [Google Scholar] [CrossRef]
  29. Fernández-Montes, A.; Velasco, F.; Ortega, J. Evaluating decision-making performance in a grid-computing environment using DEA. Expert Syst. Appl. 2012, 39, 12061–12070. [Google Scholar] [CrossRef]
  30. Campos, M.; Fernández-Montes, A.; Gavilan, J.; Velasco, F. Public resource usage in health systems: A data envelopment analysis of the efficiency of health systems of autonomous communities in Spain. Public Health 2016, 138, 33–40. [Google Scholar] [CrossRef] [PubMed]
  31. Fernández-Serrano, J.; Berbegal, V.; Velasco, F.; Expósito, A. Efficient entrepreneurial culture: A cross-country analysis of developed countries. Int. Entrep. Manag. J. 2017, 14, 105–127. [Google Scholar] [CrossRef]
  32. Exposito, A.; Velasco, F. Municipal solid-waste recycling market and the European 2020 Horizon Strategy: A regional efficiency analysis in Spain. J. Clean. Prod. 2018, 172, 938–948. [Google Scholar] [CrossRef]
  33. Scheel, H. Undesirable outputs in efficiency valuations. Eur. J Oper. Res. 2001, 132, 400–410. [Google Scholar] [CrossRef]
  34. Färe, R.; Grosskopf, S.; Hernandez-Sancho, F. Environmental performance: An index number approach. Resour. Energy Econ. 2004, 26, 343–352. [Google Scholar] [CrossRef]
  35. Zhou, P.; Ang, B.W.; Poh, K.L. Measuring environmental performance under different environmental DEA technologies. Energy Econ. 2008, 30, 1–14. [Google Scholar] [CrossRef]
  36. Sueyoshi, T.; Goto, M. Returns to scale and damages to scale on US fossil fuel power plants: Radial and non-radial approaches for DEA environmental assessment. Energy Econ. 2012, 34, 2240–2259. [Google Scholar] [CrossRef]
  37. Sueyoshi, T.; Goto, M. DEA radial measurement for environmental assessment: A comparative study between Japanese chemical and pharmaceutical firms. Appl. Energy 2014, 115, 502–513. [Google Scholar] [CrossRef]
  38. Halkos, G.E.; Tzeremes, N.G. Measuring the effect of Kyoto protocol agreement on countries’ environmental efficiency in CO2 emissions: An application of conditional full frontiers. J. Prod. Anal. 2014, 41, 367–382. [Google Scholar] [CrossRef][Green Version]
  39. Sanz-Díaz, M.T.; Velasco-Morente, F.; Yñiguez, R.; Díaz-Calleja, E. An analysis of Spain’s global and environmental efficiency from a European Union perspective. Energy Policy 2017, 104, 183–193. [Google Scholar] [CrossRef]
  40. Vlontzos, G.; Pardalos, P. Assess and prognosticate green house gas emissions from agricultural production of EU countries, by implementing, DEA Window analysis and artificial neural networks. Renew. Sustain. Energy Rev. 2017, 76, 155–162. [Google Scholar] [CrossRef]
  41. Yu, S.H.; Gao, Y.; Shiue, Y.C. A Comprehensive Evaluation of Sustainable Development Ability and Pathway for Major Cities in China. Sustainability 2017, 9, 1483. [Google Scholar] [CrossRef]
  42. Dios-Palomares, R.; Alcaide, D.; Pérrez, J.D.; Bello, M.J.; Prieto, A.; Zúniga, C.A. The Environmental Efficiency using Data Envelopment Analysis: Empirical methods and evidences. In The stated of the Art for Bieconomic and Climate Change; Editorial Universitaria UNAN Leon, Ed.; Red de Bioeconomia y Cambio Climático: Cordoba, Spain, 2017; p. 48. [Google Scholar]
  43. Cooper, W.W.; Park, K.S.; Pastor, J.T. RAM: A range adjusted measure of inefficiency for use with additive models, and relations to other models and measures in DEA. J. Prod. Anal. 1999, 11, 5–42. [Google Scholar] [CrossRef]
  44. Delimitrou, C.; Kozyrakis, C. Paragon: QoS-aware scheduling for heterogeneous datacenters. In Proceedings of the Eighteenth International Conference on Architectural Support for Programming Languages and Operating Systems, Houston, TX, USA, 16–20 March 2013; pp. 77–88. [Google Scholar]
  45. Isard, M.; Prabhakaran, V.; Currey, J.; Wieder, U.; Talwar, K.; Goldberg, A. Quincy: Fair scheduling for distributed computing clusters. In Proceedings of the ACM SIGOPS 22nd Symposium on Operating Systems Principles, Big Sky, MT, USA, 11–14 October 2009; pp. 261–276. [Google Scholar]
  46. Delimitrou, C.; Sanchez, D.; Kozyrakis, C. Tarcil: Reconciling scheduling speed and quality in large shared clusters. In Proceedings of the Sixth ACM Symposium on Cloud Computing, Kohala Coast, HI, USA, 27–29 August 2015; pp. 97–110. [Google Scholar]
  47. Grandl, R.; Ananthanarayanan, G.; Kandula, S.; Rao, S.; Akella, A. Multi-resource packing for cluster schedulers. ACM SIGCOMM Comput. Commun. Rev. 2015, 44, 455–466. [Google Scholar] [CrossRef]
  48. Zaharia, M.; Borthakur, D.; Sen Sarma, J.; Elmeleegy, K.; Shenker, S.; Stoica, I. Delay scheduling: A simple technique for achieving locality and fairness in cluster scheduling. In Proceedings of the 5th European Conference on Computer systems, Paris, France, 13–16 April 2010; pp. 265–278. [Google Scholar]
  49. Delimitrou, C.; Kozyrakis, C. Quasar: Resource-efficient and QoS-aware cluster management. In Proceedings of the 19th International Conference on Architectural Support for Programming Languages and Operating Systems, Salt Lake City, UT, USA, 1–5 March 2014; pp. 127–144. [Google Scholar]
  50. Zhang, X.; Tune, E.; Hagmann, R.; Jnagal, R.; Gokhale, V.; Wilkes, J. CPI 2: CPU performance isolation for shared compute clusters. In Proceedings of the 8th ACM European Conference on Computer Systems, Prague, The Czech Republic, 15–17 April 2013; pp. 379–391. [Google Scholar]
  51. Verma, A.; Pedrosa, L.; Korupolu, M.; Oppenheimer, D.; Tune, E.; Wilkes, J. Large-scale cluster management at Google with Borg. In Proceedings of the Tenth European Conference on Computer Systems, Bordeaux, France, 21–24 April 2015; p. 18. [Google Scholar] [CrossRef]
  52. Fernández-Cerero, D.; Fernández-Montes, A.; Jakóbik, A.; Kołodziej, J.; Toro, M. SCORE: Simulator for cloud optimization of resources and energy consumption. Simul. Model. Pract. Th. 2018, 82, 160–173. [Google Scholar] [CrossRef]
  53. Hindman, B.; Konwinski, A.; Zaharia, M.; Ghodsi, A.; Joseph, A.D.; Katz, R.H.; Shenker, S.; Stoica, I. Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center. In Proceedings of the 8th USENIX Conference on Networked Systems Design and Implementation, Boston, MA, USA, 30 March–1 April 2011; pp. 295–308. [Google Scholar]
  54. Vavilapalli, V.K.; Murthy, A.C.; Douglas, C.; Agarwal, S.; Konar, M.; Evans, R.; Graves, T.; Lowe, J.; Shah, H.; Seth, S.; et al. Apache hadoop yarn: Yet another resource negotiator. In Proceedings of the 4th Annual Symposium on Cloud Computing, Santa Clara, CA, USA, 1–3 October 2013; p. 5. [Google Scholar]
  55. Schwarzkopf, M.; Konwinski, A.; Abd-El-Malek, M.; Wilkes, J. Omega: Flexible, scalable schedulers for large compute clusters. In Proceedings of the 8th ACM European Conference on Computer Systems, Prague, The Czech Republic, 15–17 April 2013; pp. 351–364. [Google Scholar]
  56. Armbrust, M.; Fox, A.; Griffith, R.; Joseph, A.D.; Katz, R.; Konwinski, A.; Lee, G.; Patterson, D.; Rabkin, A.; Stoica, I.; et al. A view of cloud computing. Commun. ACM 2010, 53, 50–58. [Google Scholar] [CrossRef]
  57. Ruth, S. Reducing ICT-related carbon emissions: An exemplar for global energy policy? IETE Tech. Rev. 2011, 28, 207–211. [Google Scholar] [CrossRef]
Figure 1. Monolithic scheduler architecture. M, worker node; S, service task; B, batch task [52].
Figure 1. Monolithic scheduler architecture. M, worker node; S, service task; B, batch task [52].
Energies 11 02053 g001
Figure 2. Two-level scheduler architecture. C, commit; O, resource offer; SA-, scheduler agent [52].
Figure 2. Two-level scheduler architecture. C, commit; O, resource offer; SA-, scheduler agent [52].
Energies 11 02053 g002
Figure 3. Shared-state scheduler architecture. U, cluster state update [52].
Figure 3. Shared-state scheduler architecture. U, cluster state update [52].
Energies 11 02053 g003
Figure 4. Methodology workflow employed in this work. DEA, data envelopment analysis.
Figure 4. Methodology workflow employed in this work. DEA, data envelopment analysis.
Energies 11 02053 g004
Figure 5. Machine power states [52].
Figure 5. Machine power states [52].
Energies 11 02053 g005
Figure 6. Summary of DEA natural constant returns to scale (CRS) efficiency results for energy efficiency policies.
Figure 6. Summary of DEA natural constant returns to scale (CRS) efficiency results for energy efficiency policies.
Energies 11 02053 g006
Table 1. DEA inputs and outputs. Action column arrows mean whether the input/output value may be decreased (down arrow), increased (up arrow) or kept equal.
Table 1. DEA inputs and outputs. Action column arrows mean whether the input/output value may be decreased (down arrow), increased (up arrow) or kept equal.
Data-center sizeNumber of machines in the data-center↓ ↔
#shut-downsNumber of shut-down operations↓ ↔
Computation timeTotal amount of useful task computation↑ ↔
Energy consumptionTotal data-center energy consumption↓ ↔
Queue timeAverage time until jobs are fully scheduled↓ ↔
Table 2. Sample from the dataset for DEA analysis. The full dataset showing the results for the 108 DMUs analyzed can be found as the Supplementary Material. Energy policies, scheduling models, data-center sizes and workload types can be found in Section 3, Section 4 and Section 5.2, respectively. D.C., data-center.
Table 2. Sample from the dataset for DEA analysis. The full dataset showing the results for the 108 DMUs analyzed can be found as the Supplementary Material. Energy policies, scheduling models, data-center sizes and workload types can be found in Section 3, Section 4 and Section 5.2, respectively. D.C., data-center.
DMU Inputs Outputs
Energy PolicyScheduling ModelWork-LoadD.C. Size#Shut-DownsComputing Time (h)MWh ConsumedQueue Time (ms)
Table 3. Efficiency analysis for data-center sizes.
Table 3. Efficiency analysis for data-center sizes.
SchedulingWorkloadData-Center SizeEfficiency
ModelScenario1000500010,000 σ x ¯
x ¯ 0.990.440.23
Table 4. Efficiency analysis of energy policies.
Table 4. Efficiency analysis of energy policies.
Scheduling Model
Policy1000500010,0001000500010,0001000500010,000 σ x ¯
Table 5. Resulting proposed corrections following DEA analysis. Peer projections for a DMU indicate which DMU it should emulate. The following actions may be taken for each input and output: ↑ when the parameter must be increased; ↓ if the parameter must be reduced; and ↔ if no further actions are needed to achieve efficiency.
Table 5. Resulting proposed corrections following DEA analysis. Peer projections for a DMU indicate which DMU it should emulate. The following actions may be taken for each input and output: ↑ when the parameter must be increased; ↓ if the parameter must be reduced; and ↔ if no further actions are needed to achieve efficiency.
10MarginMesosHigh4 (88%)
29Exp.MesosLow23 (56%)
22 (48%)
33LoadOmegaLow31 (100%)
42GammaMono.High6 (59%)
41 (41%)
44RandomMesosHigh7 (100%)
53Exp.OmegaHigh16 (63%)
52 (36%)
70MarginOmegaLow18 (72%)
76MarginMono.High6 (55%)
40 (45%)
78GammaMono.High6 (90%)
88MarginOmegaHigh18 (95%)
90GammaOmegaHigh18 (96%)
102GammaMesosLow1 (49%)
22 (38%)
104RandomOmegaLow13 (53%)
34 (36%)
Table 6. Corrections proposed for DMU #104.
Table 6. Corrections proposed for DMU #104.
Results for DMU #104
Natural Efficiency = 0.1697
Projection Summary:
OutputComputation (h)46.09+83%084.37
OutputMWh consumed233.50−83%039.62
OutputQueue time (ms)3.80−83%00.6
Listing of Peers:
PeerLambda Weight

Share and Cite

MDPI and ACS Style

Fernández-Cerero, D.; Fernández-Montes, A.; Velasco, F. Productive Efficiency of Energy-Aware Data Centers. Energies 2018, 11, 2053.

AMA Style

Fernández-Cerero D, Fernández-Montes A, Velasco F. Productive Efficiency of Energy-Aware Data Centers. Energies. 2018; 11(8):2053.

Chicago/Turabian Style

Fernández-Cerero, Damián, Alejandro Fernández-Montes, and Francisco Velasco. 2018. "Productive Efficiency of Energy-Aware Data Centers" Energies 11, no. 8: 2053.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop