Data Center Power Research Articles

A Comparative Analysis of Task Scheduling Algorithms of Virtual Machines in Cloud Environment

Cloud computing is an interesting and beneficial area in modern distributed computing. It enables millions of users to use the offered services through their own devices or terminals. Cloud computing offers an environment with low cost, ease of use and low power consumption by utilizing server virtualization in its offered services (e.g., Infrastructure as a Service). The pool of Virtual Machines (VMs) in a cloud computing Data Center (DC) needs to be managed through an efficient task scheduling algorithm to maintain quality of service and resource utilization and thus ensure the positive impact of energy consumption in the cloud computing environment. In this study, an experimental comparative study is carried out among three task scheduling algorithms in cloud computing, namely, random resource selection, round robin and green scheduler. Based on the analysis of the simulation result, we can conclude which algorithm is the best for scheduling in terms of energy and performance of VMs. The evaluation of these algorithms is based on three metrics: Total power consumption, DC load and VM load. A number of experiments with various aims are completed in this empirical comparative study. The results showed that there is no algorithm that is superior to the others. Each has its own pros and cons. Based on the simulation performed, the green scheduler gives the best performance with respect to energy consumption. On the other hand, the random scheduler showed the best performance with respect to both VM and DC load. The round robin scheduler gives better VM and DC load than the green scheduler but have more energy consumption than both random and green schedulers. However, since the RR scheduler distributes the tasks fairly, the network traffic is balanced and neither the server nor the network node will get overloaded or congested.

Energy Efficient Multi-Level Network Resources Management in Cloud Computing Data Centers

Energy efficiency is a major issue in Cloud computing infrastructure. The large power consumption is mainly attributed to the large number of modern data centers operating within. Developing these data centers includes dynamically expanding their infrastructures to meet the ever-increasing demand for huge computation, large storage, and massive communication. Energy conservation through optimization of resources and management policies in the Cloud are a viable solution. Using virtualization to save power and employing such practices as using Virtual Machines (VMs), Server Consolidation, and VM Live Migration. This paper investigates the opportunities for Green Cloud Computing (GCC) to obtain a more comprehensive prospect towards achieving energy efficient Cloud Computing, and presents an energy efficient network resources management approach in an Infrastructure as a Service (IaaS) Cloud model. We focus on developing an energy efficient algorithm by proposing a practical multi-level Cloud Resource-Network Management (CRNM) algorithm, which is implemented in a virtual Cloud environment using Snooze framework as the Cloud energy efficiency manager. The optimization focuses on the utilization of network bandwidth as main resource under test, while also taking into account the other resources, such as CPU and memory, to get the desired performance. We choose a fat tree topology as a common three tier architecture for Cloud data canters. We conclude that our proposed algorithm will save up to 75% of power consumption in Cloud data centers, with an observed increase in efficiency compared to Non-Power Aware (NPA), Power aware(PA), and Greedy algorithms, where network elements consume about 30% of the total power of Cloud data centers.

Blocking probability in optical interconnects in data center networks

Cloud computing and Web-based applications are creating a need for powerful data centers. Data centers have a great need for high bandwidth, low latency, low blocking probability, and low bit-error rate to sustain the interaction between different applications. Current data center networks (DCNs) suffer from several problems such as high-energy consumption, high latency, fixed throughput of links, and limited reconfigurability. Electronic switches are low radix and have high latency due to a large hop count since each hop employs a store-and-forward mechanism. Optical interconnects, on the other hand, offer several advantages such as low-energy consumption, high bandwidth, reconfigurability, malleability to changing traffic, high-radix switch design, fast switching transition times, and wavelength multiplexing. These benefits provide the incentive to shift from electrical interconnects to optical interconnects in DCNs. Despite several advantages over their electrical counterparts, the performance of optical interconnects can be further improved by considering some performance parameters of optical interconnects. One such important parameter for the performance of any communication network is the blocking probability. This paper makes a comprehensive investigation of the performance of optical interconnects in different DCN architectures on the basis of blocking probability and concludes by suggesting ways to reduce the blocking.

Paralleled Three-Phase Current-Source Rectifiers for High-Efficiency Power Supply Applications

This paper presents the paralleling operation of three-phase current-source rectifiers (CSRs) as the front-end power conversion stage of data center power supply systems based on 400- $\hbox{V}_{\rm dc} $ power delivery architecture, which has been proven to have higher efficiency than traditional ac architectures. A control algorithm of paralleled three-phase CSRs is introduced to achieve balanced outputs and individual rectifier module hot swap, which are required by power supply systems. By using silicon carbide (SiC) power semiconductors, SiC MOSFETs, and Schottky diodes, the power losses of the front-end stage are reduced, and the power supply system efficiency can be further increased. The prototype of a 19-kW front-end rectifier to convert 480 $\hbox{V}_{{\rm ac},{\rm rms}}$ to 400 $\hbox{V}_{\rm dc} $ , based on three paralleled three-phase CSRs, is developed. Each CSR is an all-SiC converter and designed for high efficiency, and the front-end stage full-load efficiency is greater than 98% from experimental tests. The balanced outputs and individual converter hot swap are realized in the hardware prototype too.

A Novel Load Adaptive ZVS Auxiliary Circuit for PWM Three-Level DC–DC Converters

Three-level PWM dc–dc converters convert high dc voltage (>500 V) generally at the output of a three-phase ac–dc PWM rectifier in ac–dc converters to an isolated dc output voltage which can be used to power data center loads. Strict efficiency requirements at loads from 20% to 50% of full load of ac–dc converters for telecom applications have been introduced by energy star enforcing industries to improve efficiency of the dc–dc converter in an ac–dc converter powering data-center loads at those loads. High-efficiency requirements at low and mid loads in high switching frequency PWM dc–dc three-level converters implemented with MOSFETs can be achieved by reducing switching losses through optimized load adaptive ZVS for the entire load range. In this paper, a simple yet novel load adaptive ZVS auxiliary circuit for three-level converter is proposed for so that the resulting three-phase ac–dc converter can meet energy star platinum efficiency standard. The operation of the proposed dc–dc converter is described, analyzed, and validated by experimental results from an industrial prototype of a three-phase ac–dc converter comprising of a front-end three-phase boost PWM rectifier followed by the proposed converter.

Reducing Power Consumption in Data Centers by Jointly Considering VM Placement and Flow Scheduling

Two important components that consume the majority of IT power in data centers are the servers and the Data Center Network (DCN). Existing works fail to fully utilize power management techniques on the servers and in the DCN at the same time. In this paper, we jointly consider VM placement on servers with scalable frequencies and flow scheduling in the DCN, to minimize the overall system’s power consumption. Due to the convex relation between a server’s power consumption and its operating frequency, we prove that, given the number of servers to be used, computation workloads should be allocated to severs in a balanced way, to minimize the power consumption on servers. To reduce the power consumption of the DCN, we further consider the flow requirements among the VMs during VM allocation and assignment. Also, after VM placement, flow consolidation is conducted to reduce the number of active switches and ports. We notice that, choosing the minimum number of servers to accommodate the VMs may result in high power consumption on servers, due to servers’ increased operating frequencies. Choosing the optimal number of servers purely based on servers’ power consumption leads to reduced power consumption on servers, but may increase power consumption of the DCN. We propose to choose the optimal number of servers to be used, based on the overall system’s power consumption. Simulations show that, our joint power optimization method helps to reduce the overall power consumption significantly, and outperforms various existing state-of-the-art methods in terms of reducing the overall system’s power consumption.

Energy Efficient Virtual Network Embedding for Cloud Networks

Network virtualization is widely considered to be one of the main paradigms for the future Internet architecture as it provides a number of advantages including scalability, on demand allocation of network resources, and the promise of efficient use of network resources. In this paper, we propose an energy efficient virtual network embedding (EEVNE) approach for cloud computing networks, where power savings are introduced by consolidating resources in the network and data centers. We model our approach in an IP over WDM network using mixed integer linear programming (MILP). The performance of the EEVNE approach is compared with two approaches from the literature: the bandwidth cost approach (CostVNE) and the energy aware approach (VNE-EA). The CostVNE approach optimizes the use of available bandwidth, while the VNE-EA approach minimizes the power consumption by reducing the number of activated nodes and links without taking into account the granular power consumption of the data centers and the different network devices. The results show that the EEVNE model achieves a maximum power saving of 60% (average 20%) compared to the CostVNE model under an energy inefficient data center power profile. We develop a heuristic, real-time energy optimized VNE (REOViNE), with power savings approaching those of the EEVNE model. We also compare the different approaches adopting an energy efficient data center power profile. Furthermore, we study the impact of delay and node location constraints on the energy efficiency of virtual network embedding. We also show how VNE can impact the design of optimally located data centers for minimal power consumption in cloud networks. Finally, we examine the power savings and spectral efficiency benefits that VNE offers in optical orthogonal division multiplexing networks.

A Stackelberg Game-Based Optimization Framework of the Smart Grid With Distributed PV Power Generations and Data Centers

The emergence of cloud computing has established a trend toward building massive, energy-hungry, and geographically distributed data centers. Due to their enormous energy consumption, data centers are expected to have a major impact on the electric power grid by significantly increasing the load at locations where they are built. Dynamic energy pricing policies in the recently proposed smart power grid technology can incentivize the cloud controller to shift the computation load toward data centers in regions with cheaper electricity or with excessive electricity generated by renewable energy sources, e.g., photovoltaic (PV) and wind power. On the other hand, distributed data centers in the cloud also provide opportunities to help the power grid with distributed renewable energy sources to improve robustness and load balancing. To shed some light into these opportunities, this paper considers an interaction system of the smart power grid with distributed PV power generation and the cloud computing system, jointly accounting for the service request dispatch and routing problem in the cloud with the power flow analysis in power grid. The Stackelberg (sequential) game formulation is provided for the interaction system under two different dynamic pricing scenarios: 1) real-time power-dependent pricing; and 2) time-ahead pricing. The two players in the Stackelberg games are the power grid controller that sets the pricing signal and the cloud controller that performs resource allocation among data centers. The objective of the power grid controller is to maximize its own profit and perform load balancing among power buses, i.e., minimizing the power flow from one power bus to the others, whereas the objective of the cloud computing controller is to maximize its own profit with respect to the location-dependent pricing signal. Based on the backward induction method, this paper derives the near-optimal or suboptimal strategies of the two players in Stackelberg game using convex optimization and simulated annealing techniques.

LofoSwitch: An online policy for concerted server and disk power control in content distribution networks

The global content distribution infrastructure is expanding towards the edge of the Internet with cache servers deployed closer to the clients to enable a highly available and bandwidth-efficient delivery of high-quality video services to a broadening range of consumer terminals. The proliferation of cache servers in telco data centers contributes to the unsustainable increase of data-center power consumption. In this paper, we propose to save energy in content distribution networks (CDNs) by means of combined dynamic power management (DPM) of the CDN’s cache servers and the disks in these servers. We present an online load-directed threshold-based power control policy that reduces the CDN’s power consumption while ensuring the best possible viewing experience and a high service availability. In addition, the proposed policy avoids an increase of the bandwidth cost and an excessive wear of the servers and disks. We compare such online policy with a near-optimal offline heuristic greedy policy that determines an upper bound for the energy savings that can be realized by DPM. The evaluation of the power control policies is based on a CDN energy simulator driven by HTTP adaptive-streaming workload traces recorded by an operational telco CDN delivering IPTV services to mobile devices. Even for a minimally provisioned CDN, the online policy reduces the cache-server power consumption by 24%. The energy savings realized by the online policy correspond to 85% of the energy-savings upper bound determined by the offline policy.

Physics-based simulation of the impact of demand response on lead-acid emergency power availability in a datacenter

This paper uses a one-dimensional, physics-based model of a valve-regulated lead-acid (VRLA) battery to examine the impact of demand response on uninterruptible power supply (UPS) availability in a datacenter. Datacenters are facilities that provide services such as cloud computing, web search, etc. They are also large electricity consumers. An energy-efficient 15 MW datacenter, for instance, may pay $1 m per month for electricity. Datacenters often utilize VRLA batteries to ensure high reliability in serving their computational demand. This motivates the paper's central question: to what extent does the use of datacenter UPS batteries for demand response affect their availability for their primary purpose (namely, emergency power)? We address this question using a physics-based model of the coupled diffusion-reaction dynamics of VRLA batteries. We discretize this model using finite differences, and simulate it for different datacenter battery pack sizes. The results show that for a typical datacenter power demand profile, a VRLA battery pack sized for UPS functionality can provide demand response with only a minimal loss of UPS availability.

BitTorrent Content Distribution in Optical Networks

In this paper, we extend our previous study on BitTorrent, the most popular peer-to-peer (P2P) protocol, to investigate different aspects related to its energy efficiency in IP over WDM (IP/WDM) networks, validating the power savings previously obtained by modeling and simulation through experimental results. Our contributions can be summarized as follows: First, we compare the energy consumption of our previously proposed energy efficient BitTorrent protocol to that of the original BitTorrent protocol and the client-server (C-S) schemes over bypass IP/WDM networks considering a range of network topologies with different number of nodes and average hop counts. Our results show that for a certain swarm size, the energy efficient BitTorrent protocol achieves higher power savings in networks with lower number of nodes as the opportunity to localize traffic increases. Second, we extend our previously developed energy efficient BitTorrent heuristic enhancing its performance by allowing peers to progressively traverse more hops in the network if the number of peers in the local node is not sufficient. Third, we extend our previously developed mixed integer linear programming model to optimize the location as well as the upload rates of operator controlled seeders (OCS) to mitigate the performance degradation caused by leechers leaving after finishing the downloading operation. Fourth, we compare the power consumption of video on demand (VoD) services delivered using content distribution networks (CDN), P2P, and a promising hybrid CDN-P2P architecture over bypass IP/WDM core networks. A MILP model is developed to carry out the comparison. We investigate two scenarios for the hybrid CDN-P2P architecture: the H-MinNPC model where the model minimizes the IP/WDM network power consumption and the H-MinTPC model where the model minimizes the total power consumption including the network and the CDN datacenters power consumption. Finally, we carry out an experimental evaluation of the original and energy efficient BitTorrent heuristics.

Building a network highway for big data: architecture and challenges

Big data, with their promise to discover valuable insights for better decision making, have recently attracted significant interest from both academia and industry. Voluminous data are generated from a variety of users and devices, and are to be stored and processed in powerful data centers. As such, there is a strong demand for building an unimpeded network infrastructure to gather geologically distributed and rapidly generated data, and move them to data centers for effective knowledge discovery. The express network should also be seamlessly extended to interconnect multiple data centers as well as interconnect the server nodes within a data center. In this article, we take a close look at the unique challenges in building such a network infrastructure for big data. Our study covers each and every segment in this network highway: the access networks that connect data sources, the Internet backbone that bridges them to remote data centers, as well as the dedicated network among data centers and within a data center. We also present two case studies of real-world big data applications that are empowered by networking, highlighting interesting and promising future research directions.

Analysis of Data Center Power System

Data center presented in this paper, has a high requirement on the power supply system, that needs emergency power supply, SVG compensation device, large diesel generator set, two lines of mains large capacity UPS and other electric equipments. And there are some differences between data center and conventional power system in many aspects,such as in the nature, quantity, the load requirements, the related electrical equipment configuration, load, power supply mode, connection,and so on.

Customer satisfaction‐aware scheduling for utility maximization on geo‐distributed data centers

SummaryWith the increasingly growing amount of service requests from the world‐wide customers, the cloud systems are capable of providing services while meeting the customers' satisfaction. Recently, to achieve the better reliability and performance, the cloud systems have been largely depending on the geographically distributed data centers. Nevertheless, the dollar cost of service placement by service providers (SP) differ from the multiple regions. Accordingly, it is crucial to design a request dispatching and resource allocation algorithm to maximize net profit. The existing algorithms are either built upon energy‐efficient schemes alone, or multi‐type requests and customer satisfaction oblivious. They cannot be applied to multi‐type requests and customer satisfaction‐aware algorithm design with the objective of maximizing net profit. This paper proposes an ant‐colony optimization‐based algorithm for maximizing SP's net profit (AMP) on geographically distributed data centers with the consideration of customer satisfaction. First, using model of customer satisfaction, we formulate the utility (or net profit) maximization issue as an optimization problem under the constraints of customer satisfaction and data centers. Second, we analyze the complexity of the optimal requests dispatchment problem and rigidly prove that it is an NP‐complete problem. Third, to evaluate the proposed algorithm, we have conducted the comprehensive simulation and compared with the other state‐of‐the‐art algorithms. Also, we extend our work to consider the data center's power usage effectiveness. It has been shown that AMP maximizes SP net profit by dispatching service requests to the proper data centers and generating the appropriate amount of virtual machines to meet customer satisfaction. Moreover, we also demonstrate the effectiveness of our approach when it accommodates the impacts of dynamically arrived heavy workload, various evaporation rate and consideration of power usage effectiveness. Copyright © 2014 John Wiley & Sons, Ltd.

Research on Applying SDN Technology to Electric Power Data Center

In order to meet the developing demand of the new generation of data center, and to better satisfy the need of the smart grid, we introduce the OpenFlow-based SDN technology to electric power data center. Because there are many switches in electric power data center, it requires the use of a distributed multi-controller to manage, however there is currently no mechanism for these controllers to cooperate with the each other. Based on these questions this paper presents a collaborative model for multi-domains OpenFlow networks, which allows different network domains to collaborate with each other in an efficient way.

No more electrical infrastructure

We consider the use of fuel cells for powering data centers, based on benefits in reliability, capital and operational costs, and reduced environmental emissions. Using fuel cells effectively in data centers introduces several challenges and we highlight key research questions for designing a fuel cell based data center power distribution system. We analyze a specific configuration in the design space to quantify the cost benefits for a large scale data center, for the most mature and commonly deployed fuel cell technology, achieving over 20% reduction in costs using conservative projections.

BlinkFS: A distributed file system for intermittent power

The ability to use intermittent power in data centers introduces numerous new opportunities for optimization, including (i) using real-time markets to buy more power when it is cheap, (ii) increasing the fraction of clean, but intermittent, renewable power, (iii) capping power for long periods to extend UPS lifetime during blackouts, and (iv) fully utilizing a data center's power delivery infrastructure. The capability to run off intermittent power also moves us closer to the vision of a net-zero data center that consumes no net energy from the electric grid and has a small carbon footprint. However, designing systems to operate under intermittent power is challenging, since applications often access persistent distributed state, where power fluctuations can impact data availability and I/O performance. To address the problem, we design and implement BlinkFS, which combines blinking with a power-balanced data layout and popularity-based replication/reclamation to optimize I/O throughput and latency as power varies. Our experiments show that BlinkFS outperforms approaches that co-opt existing energy-proportional distributed file systems (DFSs) for intermittent power, particularly at low steady power levels and high levels of intermittency. For example, we show that BlinkFS reduces completion time for MapReduce-style jobs by 42% at 50% full power compared to existing energy-proportional DFSs.

Maximizing the revenues of data centers in regulation market by coordinating with electric vehicles

Frequency regulation is a major market service to reduce the undesired imbalance between power supply and demand in the power market. In order to participate in the regulation market, both the supply and demand sides need to be capable of flexibly adjusting their power generation and consumption, respectively. As the scale of Internet data centers is increasing rapidly, their significant power consumption has enabled them to become an important player in the regulation market for maximized profits and thus minimized operating expenses. On the other side, Plug-in Hybrid Electric Vehicles (PHEVs) have also recently been identified as a major participant in the regulation market, due to their large power demand for battery charging.In this paper, we propose a novel power management scheme that jointly leverages a data center and its employees’ PHEVs to (1) maximize the revenues that the data center receives from the regulation market and (2) get the PHEVs charged at no expense to their owners. Our scheme features a two-level hierarchical power control design. At the first level, our scheme interacts with the regulation market to provide information about the data center power consumption on an hourly basis. At the second level, the scheme decides the power budgets for the servers and UPS in the data center, as well as PHEVs, in real time, to follow the given regulation signal. In addition, we show how to leverage the thermal energy storage (TES) tanks available in many data centers to adapt the cooling power consumption for better management of the data center power demand and further increased regulation revenues. We evaluate the proposed scheme with real-world workload and regulation traces. The results show that our scheme performs a high-quality regulation service. As a result, the proposed scheme outperforms several commonly used baselines by having higher regulation revenues, and so lower operating expenses, for the data center. Finally, we analyze the cost savings of the PHEV owners, throughout the lifetime of the PHEVs, by getting their batteries charged at no expense.

Exploring smart grid and data center interactions for electric power load balancing

The operation of a data center consumes a tremendous amount of electricity, and the energy cost accounts for a large portion of the data center's operation cost. This leads to a growing interest towards reducing the energy cost of data centers. One approach advocated in recent studies is to distribute the computation workload among multiple geographically dispersed data centers by exploiting the electricity price differences. However, the impact of load redistributions on the power grid is not well understood yet. This paper takes the first step towards tackling this important issue, by studying how the power grid can take advantage of the data center's load distribution proactively for the purpose of power load balancing. We model the interactions between power grid and data centers as a two-stage problem, where the power grid operator aims to balance the electric power load in the first stage, and the data centers seek to minimize their total energy cost in the second stage. We show that this two-stage problem is a bilevel program with an indefinite quadratic objective function, which cannot be solved efficiently using standard convex optimization algorithms. Therefore, we reformulate this bilevel optimization problem as a linear program with additional finite complementarity slackness conditions, and propose a branch and bound algorithm to attain the globally optimal solution. The simulation results demonstrate that our proposed scheme can improve the load balancing performance by around 12% in terms of the electric load index and reduce the energy cost of data centers by 46% on average.

An Integrated Modeling Approach to Evaluate and Optimize Data Center Sustainability, Dependability and Cost

Data centers have evolved dramatically in recent years, due to the advent of social networking services, e-commerce and cloud computing. The conflicting requirements are the high availability levels demanded against the low sustainability impact and cost values. The approaches that evaluate and optimize these requirements are essential to support designers of data center architectures. Our work aims to propose an integrated approach to estimate and optimize these issues with the support of the developed environment, Mercury. Mercury is a tool for dependability, performance and energy flow evaluation. The tool supports reliability block diagrams (RBD), stochastic Petri nets (SPNs), continuous-time Markov chains (CTMC) and energy flow (EFM) models. The EFM verifies the energy flow on data center architectures, taking into account the energy efficiency and power capacity that each device can provide (assuming power systems) or extract (considering cooling components). The EFM also estimates the sustainability impact and cost issues of data center architectures. Additionally, a methodology is also considered to support the modeling, evaluation and optimization processes. Two case studies are presented to illustrate the adopted methodology on data center power systems.