An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

https://beloglazov.info/papers/2011-energy-aware-fgcs.pdf

Energy-aware resource allocation heuristics for efficient management of data centers for Cloud computing

The rapid growth in demand for computational power driven by modern service applications combined with the shift to the Cloud computing model have led to the establishment of large-scale virtualized data centers. Such data centers consume enormous amounts of electrical energy resulting in high operating costs and carbon dioxide emissions. Dynamic consolidation of virtual machines (VMs) using live migration and switching idle nodes to the sleep mode allows Cloud providers to optimize resource usage and reduce energy consumption. However, the obligation of providing high quality of service to customers leads to the necessity in dealing with the energy-performance trade-off, as aggressive consolidation may lead to performance degradation. Because of the variability of workloads experienced by modern applications, the VM placement should be optimized continuously in an online manner. To understand the implications of the online nature of the problem, we conduct a competitive analysis and prove competitive ratios of optimal online deterministic algorithms for the single VM migration and dynamic VM consolidation problems. Furthermore, we propose novel adaptive heuristics for dynamic consolidation of VMs based on an analysis of historical data from the resource usage by VMs. The proposed algorithms significantly reduce energy consumption, while ensuring a high level of adherence to the service level agreement. We validate the high efficiency of the proposed algorithms by extensive simulations using real-world workload traces from more than a thousand PlanetLab VMs. Copyright © 2011 John Wiley & Sons, Ltd.

Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in Cloud data centers

Cloud computing is an emerging commercial infrastructure paradigm that promises to eliminate the need for maintaining expensive computing facilities by companies and institutes alike. Through the use of virtualization and resource time sharing, clouds serve with a single set of physical resources a large user base with different needs. Thus, clouds have the potential to provide to their owners the benefits of an economy of scale and, at the same time, become an alternative for scientists to clusters, grids, and parallel production environments. However, the current commercial clouds have been built to support web and small database workloads, which are very different from typical scientific computing workloads. Moreover, the use of virtualization and resource time sharing may introduce significant performance penalties for the demanding scientific computing workloads. In this work, we analyze the performance of cloud computing services for scientific computing workloads. We quantify the presence in real scientific computing workloads of Many-Task Computing (MTC) users, that is, of users who employ loosely coupled applications comprising many tasks to achieve their scientific goals. Then, we perform an empirical evaluation of the performance of four commercial cloud computing services including Amazon EC2, which is currently the largest commercial cloud. Last, we compare through trace-based simulation the performance characteristics and cost models of clouds and other scientific computing platforms, for general and MTC-based scientific computing workloads. Our results indicate that the current clouds need an order of magnitude in performance improvement to be useful to the scientific community, and show which improvements should be considered first to address this discrepancy between offer and demand.

/pdf/performance-analysis-of-cloud-computing-services-for-many-3e3awq09a7.pdf

Performance Analysis of Cloud Computing Services for Many-Tasks Scientific Computing

Rapid growth of the demand for computational power by scientific, business and web-applications has led to the creation of large-scale data centers consuming enormous amounts of electrical power. We propose an energy efficient resource management system for virtualized Cloud data centers that reduces operational costs and provides required Quality of Service (QoS). Energy savings are achieved by continuous consolidation of VMs according to current utilization of resources, virtual network topologies established between VMs and thermal state of computing nodes. We present first results of simulation-driven evaluation of heuristics for dynamic reallocation of VMs using live migration according to current requirements for CPU performance. The results show that the proposed technique brings substantial energy savings, while ensuring reliable QoS. This justifies further investigation and development of the proposed resource management system.

/pdf/energy-efficient-resource-management-in-virtualized-cloud-24lkelu8uu.pdf

Energy Efficient Resource Management in Virtualized Cloud Data Centers

Workload placement on servers has been traditionally driven by mainly performance objectives. In this work, we investigate the design, implementation, and evaluation of a power-aware application placement controller in the context of an environment with heterogeneous virtualized server clusters. The placement component of the application management middleware takes into account the power and migration costs in addition to the performance benefit while placing the application containers on the physical servers. The contribution of this work is two-fold: first, we present multiple ways to capture the cost-aware application placement problem that may be applied to various settings. For each formulation, we provide details on the kind of information required to solve the problems, the model assumptions, and the practicality of the assumptions on real servers. In the second part of our study, we present the pMapper architecture and placement algorithms to solve one practical formulation of the problem: minimizing power subject to a fixed performance requirement. We present comprehensive theoretical and experimental evidence to establish the efficacy of pMapper.

/pdf/pmapper-power-and-migration-cost-aware-application-placement-2o53nt8erw.pdf

pMapper: power and migration cost aware application placement in virtualized systems

High Performance Computing applications and platforms have been typically designed without regard to power consumption. With increased awareness of energy cost, power management is now an issue even for compute-intensive server clusters. In this work, we investigate the use of power management techniques for high performance applications on modern power-efficient servers with virtualization support. We consider power management techniques such as dynamic consolidation and usage of dynamic power range enabled by low power states on servers. We identify application performance isolation and virtualization overhead with multiple virtual machines as the key bottlenecks for server consolidation. We perform a comprehensive experimental study to identify the scenarios where applications are isolated from each other. We also establish that the power consumed by HPC applications may be application dependent, non-linear and have a large dynamic range. We show that for HPC applications, working set size is a key parameter to take care of while placing applications on virtualized servers. We use the insights obtained from our experimental study to present a framework and methodology for power-aware application placement for HPC applications.

/pdf/power-aware-dynamic-placement-of-hpc-applications-axz3y5okcs.pdf

Power-aware dynamic placement of HPC applications

Described are methods, apparatus and computer programs for determining run-time dependencies between logical components of a data processing environment. Components of the data processing environment are monitored by monitoring agents accessing run-time activity data via APIs of the managed system. A dependency generator identifies correlations between the run-time activity of the monitored components. For synchronous monitored systems, the dependency generator calculates an activity period for monitored components and determines which component's activity periods contain the activity periods of other components. Containment is used as an indicator of a likely dependency relationship, and a weighting is computed for each dependency relationship based on the consistency of containment.

Systems, methods and computer programs for determining dependencies between logical components in a data processing system or network

N applications are placed on M virtualized servers having power management capability. A time horizon is divided into a plurality of time windows, and, for each given one of the windows, a placement of the N applications is computed, taking into account power cost, migration cost, and performance benefit. The migration cost refers to cost to migrate from a first virtualized server to a second virtualized server for the given one of the windows. The N applications are placed onto the M virtualized servers, for each of the plurality of time windows, in accordance with the placement computed in the computing step for each of the windows. In an alternative aspect, power cost and performance benefit, but not migration cost, are taken into account; there are a plurality of virtual machines; and the computing step includes, for each of the windows, determining a target utilization for each of the servers based on a power model for each given one of the servers; picking a given one of the servers with a least power increase per unit increase in capacity, until capacity has been allocated to fit all the virtual machines; and employing a first fit decreasing bin packing technique to compute placement of the applications on the virtualized servers.

Techniques for placing applications in heterogeneous virtualized systems while minimizing power and migration cost

The increasing heterogeneity between applications in emerging virtualized data centers like clouds introduce significant challenges in estimating the power drawn by the data center. In this work, we presentWattApp: an application-aware power meter for shared data centers that addresses this challenge. In order to deal with heterogeneous applications, WattApp introduces application parameters (e.g, throughput) in the power modeling framework. WattApp is based on a carefully designed set of experiments on a mix of diverse applications: power benchmarks, web-transaction workloads, HPC workloads and I/O-intensive workloads. Given a set of N applications and M server types, WattApp runs in O(N) time, uses O(NxM) calibration runs, and predicts the power drawn by any arbitrary placement within 5%of the real power for the applications studied.

/pdf/wattapp-an-application-aware-power-meter-for-shared-data-3lxa7pg02q.pdf

Anindya Neogi

Papers

pMapper: power and migration cost aware application placement in virtualized systems

Power-aware dynamic placement of HPC applications

Systems, methods and computer programs for determining dependencies between logical components in a data processing system or network

Techniques for placing applications in heterogeneous virtualized systems while minimizing power and migration cost

WattApp: an application aware power meter for shared data centers