Energy-aware capacity scaling in virtualized environments with performance guarantees

We investigate the trade-off between performance and power consumption in servers hosting virtual machines running IT services. The performance behavior of such servers is modeled through Generalized Processor Sharing (GPS) queues enhanced with a green speed-scaling mechanism that controls the processing capacity to use depending on the number of active virtual machines. When the number of virtual machines grows large, we show that the stochastic evolution of our model converges to a system of ordinary differential equations for which we derive a closed-form formula for its unique stationary point. This point is a function of the capacity and the shares that characterize the GPS mechanism. It allows us to show that speed-scaling mechanisms can provide large reduction in power consumption having only small performance degradation in terms of the delays experienced in the virtual machines. In addition, we derive the optimal choice for the shares of the GPS discipline, which turns out to be non-trivial. Finally, we show how our asymptotic analysis can be applied to the dimensioning and service partitioning in data-centers. Experimental results show that our asymptotic formulas are accurate even when the number of virtual machines is small.     

Multiperiod robust optimization for proactive resource provisioning in virtualized data centers

Energy management has become a significant concern in data centers for reducing operational costs. Using virtualization allows server consolidation, which increases server utilization and reduces energy consumption by turning off idle servers. This needs to consider the power state change overhead. In this paper, we investigate proactive resource provisioning in short-term planning for performance and energy management. To implement short-term planning based on workload prediction, this requires dealing with high fluctuations that are inaccurately predictable by using single value prediction. Unlike long-term planning, short-term planning can not depend on periodical patterns. Thus, we propose an adaptive range-based prediction algorithm instead of a single value. We implement and extensively evaluate the proposed range-based prediction algorithm with different days of real workload. Then, we exploit the range prediction for implementing proactive provisioning using robust optimization taking into consideration uncertainty of the demand. We formulate proactive VM provisioning as a multiperiod robust optimization problem. To evaluate the proposed approach, we use several experimental setups and different days of real workload. We use two metrics: energy savings and robustness for ranking the efficiency of different scenarios. Our approach mitigates undesirable changes in the power state of servers. This enhances servers’ availability for accommodating new VMs, its robustness against uncertainty in workload change, and its reliability against a system failure due to frequent power state changes.     

Energy performance of heuristics and meta-heuristics for real-time joint resource scaling and consolidation in virtualized networked data centers

In this paper, we explore on a comparative basis the performance suitability of meta-heuristic, sometime denoted as random search algorithms, and greedy-type heuristics for the energy-saving joint dynamic scaling and consolidation of the network-plus-computing resources hosted by networked virtualized data centers when the target is the support of real-time streaming-type applications. For this purpose, the energy and delay performances of Tabu Search (TS), Simulated Annealing (SA) and Evolutionary Strategy (ES) meta-heuristics are tested and compared with the corresponding ones of Best-Fit Decreasing-type heuristics, in order to give insight on the resulting performance-versus-implementation complexity trade-offs. In principle, the considered meta-heuristics and heuristics are general formal approaches that can be applied to large classes of (typically, non-convex and mixed integer) optimization problems. However, specially for the meta-heuristics, a main challenge is to design them to properly address the real-time joint computing-plus-networking resource consolidation and scaling optimization problem. To this purpose, the aim of this paper is: (i) introduce a novel Virtual Machine Allocation (VMA) scheme that aims at choosing a suitable set of possible Virtual Machine placements among the (possibly, non-homogeneous) set of available servers; (ii) propose a new class of random search algorithms (RSAs) denoted as consolidation meta-heuristic, considering the VMA problem in RSAs. In particular, the design of novel variants of meta-heuristics, namely TS-RSC, SA-RSC and ES-RSC, is particularized to the resource scaling and consolidation (RSC) problem; (iii) compare the results of the obtained new RSAs class against some state-of-the-art heuristic approaches. A set of experimental results, both simulated and real-world ones, support the effectiveness of the proposed approaches against the traditional ones.     

Spatio-temporal thermal-aware job scheduling to minimize energy consumption in virtualized heterogeneous data centers

Job scheduling in data centers can be considered from a cyber–physical point of view, as it affects the data center’s computing performance (i.e. the cyber aspect) and energy efficiency (the physical aspect). Driven by the growing needs to green contemporary data centers, this paper uses recent technological advances in data center virtualization and proposes cyber–physical, spatio-temporal (i.e. start time and servers assigned), thermal-aware job scheduling algorithms that minimize the energy consumption of the data center under performance constraints (i.e. deadlines). Savings are possible by being able to temporally “spread” the workload, assign it to energy-efficient computing equipment, and further reduce the heat recirculation and therefore the load on the cooling systems. This paper provides three categories of thermal-aware energy-saving scheduling techniques: (a) FCFS-Backfill-XInt and FCFS-Backfill-LRH, thermal-aware job placement enhancements to the popular first-come first-serve with back-filling (FCFS-backfill) scheduling policy; (b) EDF-LRH, an online earliest deadline first scheduling algorithm with thermal-aware placement; and (c) an offline genetic algorithm for SCheduling to minimize thermal cross-INTerference (SCINT), which is suited for batch scheduling of backlogs. Simulation results, based on real job logs from the ASU Fulton HPC data center, show that the thermal-aware enhancements to FCFS-backfill achieve up to 25% savings compared to FCFS-backfill with first-fit placement, depending on the intensity of the incoming workload, while SCINT achieves up to 60% savings. The performance of EDF-LRH nears that of the offline SCINT for low loads, and it degrades to the performance of FCFS-backfill for high loads. However, EDF-LRH requires milliseconds of operation, which is significantly faster than SCINT, the latter requiring up to hours of runtime depending upon the number and size of submitted jobs. Similarly, FCFS-Backfill-LRH is much faster than FCFS-Backfill-XInt, but it achieves only part of FCFS-Backfill-XInt’s savings.     

Energy-aware framework with Markov chain-based parallel simulated annealing algorithm for dynamic management of virtual machines in cloud data centers

Significant savings in the energy consumption, without sacrificing service level agreement (SLA), are an excellent economic incentive for cloud providers. By applying efficient virtual Machine placement and consolidation algorithms, they are able to achieve these goals. In this paper, we propose a comprehensive technique for optimum energy consumption and SLA violation reduction. In the proposed approach, the issues of allocation and management of virtual machines are divided into smaller parts. In each part, new algorithms are proposed or existing algorithms have been improved. The proposed method performs all steps in distributed mode and acts in centralized mode only in the placement of virtual machines that require a global vision. For this purpose, the population-based or parallel simulated annealing (SA) algorithm is used in the Markov chain model for virtual machines placement policy. Simulation of algorithms in different scenarios in the CloudSim confirms better performance of the proposed comprehensive algorithm.     

Power and performance management for parallel computations in clouds and data centers

We address scheduling independent and precedence constrained parallel tasks on multiple homogeneous processors in a data center with dynamically variable voltage and speed as combinatorial optimization problems. We consider the problem of minimizing schedule length with energy consumption constraint and the problem of minimizing energy consumption with schedule length constraint on multiple processors. Our approach is to use level-by-level scheduling algorithms to deal with precedence constraints. We use a simple system partitioning and processor allocation scheme, which always schedules as many parallel tasks as possible for simultaneous execution. We use two heuristic algorithms for scheduling independent parallel tasks in the same level, i.e., SIMPLE and GREEDY. We adopt a two-level energy/time/power allocation scheme, namely, optimal energy/time allocation among levels of tasks and equal power supply to tasks in the same level. Our approach results in significant performance improvement compared with previous algorithms in scheduling independent and precedence constrained parallel tasks.     

EAAM: Energy-aware application management strategy for FPGA-based IoT-Cloud environments

The Journal of Supercomputing - An efficient integration of Internet of Things (IoT) and cloud computing techniques accelerates the evolution of next-generation smart environments (e.g., smart...     

Model-driven coordinated management of data centers

Management of computing infrastructure in data centers is an important and challenging problem, that needs to: (i) ensure availability of services conforming to the Service Level Agreements (SLAs); and (ii) reduce the Power Usage Effectiveness (PUE), i.e. the ratio of total power, up to half of which is attributed to data center cooling, over the computing power to service the workloads. The cooling energy consumption can be reduced by allowing higher-than-usual thermostat set temperatures while maintaining the ambient temperature in the data center room within manufacturer-specified server redline temperatures for their reliable operations. This paper proposes: (i) a Coordinated Job, Power, and Cooling Management (JPCM) policy, which performs: (a) job management so as to allow for an increase in the thermostat setting of the cooling unit while meeting the SLA requirements, (b) power management to reduce the produced thermal load, and (c) cooling management to dynamically adjust the thermostat setting; and (ii) a Model-driven coordinated Management Architecture (MMA), which uses a state-based model to dynamically decide the correct management policy to handle events, such as new workload arrival or failure of a cooling unit, that can trigger an increase in the ambient temperature. Each event is associated with a time window, referred to as the window-of-opportunity, after which the temperature at the inlet of one or more servers can go beyond the redline temperature if proper management policies are not enforced.This window-of-opportunity monotonically decreases with increase in the incoming workload. The selection of the management policy depends on their potential energy benefits and the conformance of the delays in their actuation to the window-of-opportunity. Simulations based on actual job traces from the ASU HPC data center show that the JPCM can achieve up to 18% energy-savings over separated power or job management policies. However, high delay to reach a stable ambient temperature (in case of cooling management through dynamic thermostat setting) can violate the server redline temperatures. A management decision chart is developed as part of MMA to autonomically employ the management policy with maximum energy-savings without violating the window-of-opportunity, and hence the redline temperatures. Further, a prototype of the JPCM is developed by configuring the widely used Moab cluster manager to dynamically change the server priorities for job assignment.     

GPU-aware resource management in heterogeneous cloud data centers

The Journal of Supercomputing - The power of rapid scalability and easy maintainability of cloud services is driving many high-performance computing applications from company server racks into...     

Proactive workload management in dynamic virtualized environments

Recently, with the improvement of Cloud systems technologies and the essential advantages they can provide such as availability, scalability, and costs saving; massive domains in the IT industry are directing their business to the Cloud. To fit the computing demands of this trend along with nowadays fluky applications (e.g. social networks, media contents), Cloud systems require rapid resource changes. As a result, the workload management in a virtualized environment becomes a complex task. In this paper we propose a new proactive workload management model for virtualized resources to inspect the workload behavior of the running Virtual Machines, and to assent an appropriate scheduling and resource consolidation schema in order to improve the system efficiency, utilization, and throughput. We have carried out our model by modifying Xen Cloud Platform, then we tested the model performance through different representative benchmarks. The results show that the Proactive model can decrease the average response time remarkably.     

Proactive virtualized resource management for service workflows in the cloud

A Cloud platform offers on-demand provisioning of virtualized resources and pay-per-use charge model to its hosted services to satisfy their fluctuating resource needs. Resource scaling in cloud is often carried out by specifying static rules or thresholds. As business processes and scientific jobs become more intricate and involve more components, traditional reactive or rule-based resource management methods are not able to meet the new requirements. In this paper, we extend our previous work on dynamically managing virtualized resources for service workflows in a cloud environment. Extensive experimental results of an adaptive resource management algorithm are reported. The algorithm makes resource management decisions based on predictive results and high level user specified thresholds. It is also able to coordinate resources among the component services of a workflow so that unnecessary resource allocations and terminations can be avoided. Based on observations from previous experiments, the algorithm is extended with a new resource merge strategy in order to prevent average resource size from shrinking. Simulation results from synthetic workload data demonstrated the effectiveness of the extension.     

Transparently bridging semantic gap in CPU management for virtualized environments

Consolidated environments are progressively accommodating diverse and unpredictable workloads in conjunction with virtual desktop infrastructure and cloud computing. Unpredictable workloads, however, aggravate the semantic gap between the virtual machine monitor and guest operating systems, leading to inefficient resource management. In particular, CPU management for virtual machines has a critical impact on I/O performance in cases where the virtual machine monitor is agnostic about the internal workloads of each virtual machine. This paper presents virtual machine scheduling techniques for transparently bridging the semantic gap that is a result of consolidated workloads. To enable us to achieve this goal, we ensure that the virtual machine monitor is aware of task-level I/O-boundedness inside a virtual machine using inference techniques, thereby improving I/O performance without compromising CPU fairness. In addition, we address performance anomalies arising from the indirect use of I/O devices via a driver virtual machine at the scheduling level. The proposed techniques are implemented on the Xen virtual machine monitor and evaluated with micro-benchmarks and real workloads on Linux and Windows guest operating systems.     

Utilization-based pricing for power management and profit optimization in data centers

Traditional load balancing approaches may spread the load on more computers as long as the performance in terms of response time or cost is minimized. Nowadays power is a growing cost factor for data centers. In this paper, from the service provider’s point of view, the load balancing decision is made based on whether power consumption can be reduced or more profit can be earned. To achieve this, we design pricing algorithms to influence the load distribution. Both algorithms take into account the utilization of computers besides other factors, such as prices and power costs. In the first algorithm, we design pricing functions with respect to the computer utilization to encourage or discourage resource usage. In the second algorithm, we focus on the profit that a service provider can earn after deducting power cost from its revenue. We formulate this profit optimization problem and derive the optimum price solution.     

Service level agreement based energy-efficient resource management in cloud data centers

As cloud computing has become a popular computing paradigm, many companies have begun to build increasing numbers of energy hungry data centers for hosting cloud computing applications. Thus, energy consumption is increasingly becoming a critical issue in cloud data centers. In this paper, we propose a dynamic resource management scheme which takes advantage of both dynamic voltage/frequency scaling and server consolidation to achieve energy efficiency and desired service level agreements in cloud data centers. The novelty of the proposed scheme is to integrate timing analysis, queuing theory, integer programming, and control theory techniques. Our experimental results indicate that, compared to a statically provisioned data center that runs at the maximum processor speed without utilizing the sleep state, the proposed resource management scheme can achieve up to 50.3% energy savings while satisfying response-time-based service level agreements with rapidly changing dynamic workloads.     

Energy-efficient data centers

Energy consumption of the Information and Communication Technology (ICT) sector has grown exponentially in recent years. A major component of the today’s ICT is constituted by the data centers which have experienced an unprecedented growth in their size and population, recently. The Internet giants like Google, IBM and Microsoft house large data centers for cloud computing and application hosting. Many studies, on energy consumption of data centers, point out to the need to evolve strategies for energy efficiency. Due to large-scale carbon dioxide ( $\mathrm{CO}_2$ ) emissions, in the process of electricity production, the ICT facilities are indirectly responsible for considerable amounts of green house gas emissions. Heat generated by these densely populated data centers needs large cooling units to keep temperatures within the operational range. These cooling units, obviously, escalate the total energy consumption and have their own carbon footprint. In this survey, we discuss various aspects of the energy efficiency in data centers with the added emphasis on its motivation for data centers. In addition, we discuss various research ideas, industry adopted techniques and the issues that need our immediate attention in the context of energy efficiency in data centers.     

Supporting dynamic access to virtualized data resources via WSRF-based services

Today grid applications require not only lots of computational power but data at a very large scale too. Although grid computing was initially conceptualized as the technology that focuses on solving compute-intensive problems, this focus has gradually shifted to applications where data is distributed over various locations. Access to these data resources stored in heterogeneous grid storage systems located at geographically distributed virtual organizations in an integrated and uniform way is a challenging problem. The Web Services Resource Framework (WSRF) has recently emerged as the standard for the development and integration of grid services. This paper proposes and presents Gravy4WS, a middleware architecture based on WSRF Web services that enables the dynamic access to virtualized grid data resources. A novel scheduling algorithm called DCE (Delegating-Cluster-Execution based Scheduling) is proposed to improve load balancing of the system. The implementation of Gravy4WS using WSRF libraries and services provided by Globus Toolkit 4 is described together with its performance evaluation.     

Balance of power: dynamic thermal management for Internet data centers

Internet-based applications and their resulting multitier distributed architectures have changed the focus of design for large-scale Internet computing. Internet server applications execute in a horizontally scalable topology across hundreds or thousands of commodity servers in Internet data centers. Increasing scale and power density significantly impacts the data center's thermal properties. Effective thermal management is essential to the robustness of mission-critical applications. Internet service architectures can address multisystem resource management as well as thermal management within data centers.