Power‐aware server consolidation for federated clouds

Communities that gather functionally identical or complementary cloud services aim to provide better visibility, efficiency, and market share. This paper investigates the issue of forming these communities in distributed decision-making settings under incomplete information. By incomplete information, we mean only partial information about the individual performance of cloud services within communities and about how they will behave within these communities is available. Forming communities in these particular settings is still an open problem. Most of the existing models require real-time global knowledge about the services and high computational complexity, which makes the community formation extremely hard and time-consuming. In this paper, we propose a strategic Distributed Decision-making Mechanism (DDM) that regulates the cloud services decision-making process. DDM first generates an initial set of data based on information obtained from existing cloud services regarding their single and cooperative efficiency. By analyzing this set and on the basis of a distance function, the decision-making mechanism with regard to which community to form is implemented as a decision profile of strategies and their expected utility computed in terms of computational efficiency. DDM efficiently and systematically helps 1) communities find appropriate cloud services to invite as new members and 2) single services find suitable communities to join. To evaluate the proposed mechanism, we performed experiments using real data including 142 users and 4,000 cloud services obtained from the CloudArmor, CloudHarmony, and WS-DREAM datasets. The experimental results show that our algorithms outperform the existing solutions. KEYWORDScloud services, community of services, distributed decision-making INTRODUCTIONModern applications are nowadays developed using service-oriented architecture and web services technology where the cloud's software as a service (SaaS) is de facto model for software delivery. To make these applications more efficient and adaptive to the new requirements of being flexible, cloud services including SaaS, platform as a service (PaaS), infrastructure as a service (IaaS), and backend as a service (BaaS) are shifted, from the perspective of design and engineering, from passive and individual components to intelligent, autonomous, and collaborative-based components having common goals. 1-4 Being collaborative entails new challenges towards analyzing the overall system's performance in achieving these common goals. 5-8 One of the initiatives towards strengthening this collaboration among services is capitalizing on the concept of virtual communities. 9-11 A virtual community of services is defined as a group of services offering similar and complementary functionalities with diverse non-functional properties that can ensure the inter-community and intra-community interactions.Despite the existence of the current services technologies (e.g., WSDL and SOAP) and cloud developments that enable developing virtual commun...

Section: Cloud Servicesmentioning

confidence: 99%

Dynamic formation of service communities in the cloud under distribution and incomplete information settings

Khosrowshahi-Asl¹,

Estrada

Otrok

et al. 2017

“…1-4 It considers the ideas of distributed and parallel computing to offer on-demand computing resources for computers or other devices. 15 In the last few years, the main focus of investigators and engineers was on improving the performance in terms of energy, space, and cost. [10][11][12] In the cloud data centers, scalability and complexity issues cause cost and energy become two key challenges that need attention and careful study.…”

Section: Introductionmentioning

confidence: 99%

“…However, reducing the energy consumption and also maintaining high computation capacity have become timely and important challenges. 15 In the last few years, the main focus of investigators and engineers was on improving the performance in terms of energy, space, and cost. By increasing the number of data replicas, the energy consumption, the performance, and also the cost of creating and maintaining new replicas also are increased.…”

mentioning

confidence: 99%

See 1 more Smart Citation

An energy‐aware method for data replication in the cloud environments using a Tabu search and particle swarm optimization algorithm

Ebadi

Navimipour

2018

Cloud computing is a type of parallel, configurable, and flexible system, which refers to the provision of applications on virtual data centers. However, reducing the energy consumption and also maintaining high computation capacity have become timely and important challenges. The concept of replication is used to face these challenges. By increasing the number of data replicas, the energy consumption, the performance, and also the cost of creating and maintaining new replicas also are increased. Deciding on the number of required replicas and their location on the cloud system is an NP-hard problem. In this paper, the problem is formulated as an optimization problem and a hybrid metaheuristic algorithm is offered to solve it. The algorithm uses the global search capability of the Particle Swarm Optimization (PSO) algorithm and the local search capability of the Tabu Search (TS) to get high-quality solutions. The efficiency of the method is shown by comparing it with simple PSO, TS, and Ant Colony Optimization (ACO) algorithm on different test cases. The obtained results indicate that the method outperforms all of them in terms of consumed energy and cost. KEYWORDS cloud computing, data replication, particle swarm optimization, replica INTRODUCTIONCloud computing is a deep revolution in providing the information technology-based services based on virtualization technology. 1-4 It considers the ideas of distributed and parallel computing to offer on-demand computing resources for computers or other devices. 5-9 It aims to offer shared services with high reliability, dynamicity, and scalability; although, it is facing many issues because of its complex architecture. [10][11][12] In the cloud data centers, scalability and complexity issues cause cost and energy become two key challenges that need attention and careful study. 13,14 As the carbon footprint and consumed energy of the cloud infrastructures have been increased, investigators tried to find applicable methods for decreasing energy consumption. 15 In the last few years, the main focus of investigators and engineers was on improving the performance in terms of energy, space, and cost. 1

“…Alessandro Leite et al [23] introduce a server consolidation strategy, which can greatly reduce power consumption on Cloud federations. The strategy that the authors propose aim to decrease energy consumption without affecting the provisioning of resources to match established QoS requirements.…”

mentioning

confidence: 99%

New advances in large‐scale distributed simulation and real‐time applications

Notare¹

2016

The overlap between large distributed simulations and real-time applications [1][2][3] has been always present, and it is increasing fast recently because of the circumstances raised from the current new trends, such as Cloud computing [4,5], intelligent transportation systems [1,6], green computing [7,8], and virtual environments [2,9]. As the overlap increases, new challenges appear, mostly for enabling applications, like serious gaming [10], virtual and augmented reality, and collaborative virtual environments [11], to execute seamlessly on distributed resources [12][13][14].This special issue intends to collect the state-of-the-art research works that include a set of concepts on high performance computing; design, modeling, and validation of distributed real-time systems; and multi-agent control and simulation systems. This issue also presents as objective to enable access to future trends and directions in the scope of simulations and real-time systems, covering cloud-based simulations, simulation in GPUs, and implementation of distributed virtual environment applications.Nine articles have been chosen from a strict selective process; all of these articles have been thoroughly reviewed by highly qualified anonymous referees. The articles cover a variety of important and challenging topics in the areas of large-scale distributed simulations and real-time systems.Matthew Forshaw, et. al.[15] present a new scalable simulator of High-Throughput Computing systems (HTC-Sim) incorporating real workloads and distinguishing itself by accepting to model multi-use clusters and interactive users. Their model is specially built to work with HTCSim and allow different HTC system policies to be incorporated in the simulation system. Being a large-scale simulation software, HTC-Sim provides the ability to model multi-use clusters and the presence of interactive users; thus, it enables fault tolerance and evaluation of server's energy consumption in HTC using different resource allocation policies.Arthur Valadares, Eugenia Gabrielova, and Cristina Lopes [16] introduce a conceptual framework for facilitating the design and testing of distributed real-time systems and applications. In their work, they define six properties and concerns while designing a Distributed Real Time (DRT) system. These properties are showcased in real-life, popular examples from different applicable domains in the field of distributed virtual environments. The re-evaluation of the concerns raised in this work shows the importance of some aspect in enhancing user quality of experience in such environments.Xiaosong Li, Wentong Cai, and Stephen John Turner [17] propose two major techniques in improving execution performance of Agent-based simulations on GPUs. Their innovative techniques consist of making use of an AgentPool and a location-based shared memory management. The AgentPool is introduced to support efficient agent creation and deletion and to enhance the CUDA native memory allocation for agent-based simulations. The memory management employs ...