Search

Scheduling workload is a demand response strategy for data centers to reshape electricity usage, which provides an opportunity for them to utilize renewable energy. Enhancing the flexibility of workload scheduling would promote the data center to integrate renewable energy. Considering that the improvement of flexibility in workload scheduling is tightly related to the Quality of Service (QoS) required by IT consumers (ITCs), it becomes desirable for the data center to collaborate with them to further facilitate renewable energy integration in demand response programs. This paper proposes a QoS‐based bi‐level demand response model for the data center, where game theory is adopted to resolve the conflict of interest in the collaboration between the data center and ITCs. At the upper level, the data center is a leader who provides differential monetary incentives to ITCs while dispatching workload to integrate renewable energy. At the lower level, ITCs act as followers who adjust the workload's QoS to enhance the flexibility of workload scheduling. Simulation results show that the proposed model can decrease the operating cost of the data center while increasing the utilization of renewable energy. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

The application of Artificial Intelligence (AI) and Machine Learning (ML) models is increasingly leveraged to automate and optimize Data Centre (DC) operations. However, the interpretability and transparency of these complex models pose critical challenges. Hence, this paper explores the Shapley Additive exPlanations (SHAP) values model explainability method for addressing and enhancing the critical interpretability and transparency challenges of predictive maintenance models. This method computes and assigns Shapley values for each feature, then quantifies and assesses their impact on the model's output. By quantifying the contribution of each feature, SHAP values can assist DC operators in understanding the underlying reasoning behind the model's output in order to make proactive decisions. As DC operations are dynamically changing, we additionally investigate how SHAP can capture the temporal behaviors of feature importance in the dynamic DC environment over time. We validate our approach with selected predictive models using an actual dataset from a High-Performance Computing (HPC) DC sourced from the Enea CRESCO6 cluster in Italy. The experimental analyses are formalized using summary, waterfall, force, and dependency explanations. We delve into temporal feature importance analysis to capture the features' impact on model output over time. The results demonstrate that model explainability can improve model transparency and facilitate collaboration between DC operators and AI systems, which can enhance the operational efficiency and reliability of DCs by providing a quantitative assessment of each feature's impact on the model's output. [ABSTRACT FROM AUTHOR]

Cloud computing enables highly configurable and reliable computing resources on a rentable per-use scheme, facilitating quick and cost-effective provisioning of large-scale applications. Thanks to the fast-paced evolution of cutting-edge technologies and the rapid spread of cloud-based solutions, the cloud computing ecosystem is now part of our everyday lives. Nevertheless, cloud computing relies on highly sophisticated data centers comprising energy-consuming servers and equipment that require much energy. Stimulating cloud services for active participation and network contributions presents several challenges. Strategies based on artificial intelligence (AI), game theory, and blockchain have great potential to create an economically sustainable cloud ecosystem. This paper explores strategies grounded in AI, game theory, and blockchain to foster an economically sustainable cloud ecosystem. Informed by a survey study, our research delves into incentive approaches within cloud computing. Theoretical foundations, motivations, and enabling techniques are comprehensively examined to provide valuable insights for a broad audience. The primary contributions of this work lie in elucidating the application of AI, game theory, and blockchain to address challenges in incentivizing cloud services, paving the way for a more sustainable and efficient cloud computing landscape. [ABSTRACT FROM AUTHOR]

The need for artificial intelligence (AI) and machine learning (ML) models to optimize data center (DC) operations increases as the volume of operations management data upsurges tremendously. These strategies can assist operators in better understanding their DC operations and help them make informed decisions upfront to maintain service reliability and availability. The strategies include developing models that optimize energy efficiency, identifying inefficient resource utilization and scheduling policies, and predicting outages. In addition to model hyperparameter tuning, feature subset selection (FSS) is critical for identifying relevant features for effectively modeling DC operations to provide insight into the data, optimize model performance, and reduce computational expenses. Hence, this paper introduces the Shapley Additive exPlanation (SHAP) values method, a class of additive feature attribution values for identifying relevant features that is rarely discussed in the literature. We compared its effectiveness with several commonly used, importance-based feature selection methods. The methods were tested on real DC operations data streams obtained from the ENEA CRESCO6 cluster with 20,832 cores. To demonstrate the effectiveness of SHAP compared to other methods, we selected the top ten most important features from each method, retrained the predictive models, and evaluated their performance using the MAE, RMSE, and MPAE evaluation criteria. The results presented in this paper demonstrate that the predictive models trained using features selected with the SHAP-assisted method performed well, with a lower error and a reasonable execution time compared to other methods. [ABSTRACT FROM AUTHOR]

The data center operator has diverse options to sell redundant energy generated by its on‐site renewable generators to obtain payoff, where options include the grid and surrounding consumers. Different options and combinations of options mean different transaction modes for the data center operator. However, current works mainly investigate the selling of renewable energy in a fixed transaction mode, which will lead to payoff loss of data center operator. To heighten the payoff of data center operator, this paper designs two frameworks to model two transaction modes (i.e., single‐channel transaction mode and dual‐channel transaction mode) based on the Stackelberg game, where the Stackelberg game is used to address the conflict of interest between data center operator and electric utility company. The Stackelberg equilibrium, the optimal payoff of the data center operator, in two transaction modes is derived. Based on the difference of calculated optimal payoff in two transaction modes, an algorithm is proposed to adaptively choose the transaction mode that can bring more payoff. Simulation results show that the proposed algorithm is able to increase the payoff of the data center operator compared with that in fixed mode. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Cloud computing technology provides shared computing which can be accessed over the Internet. When cloud data centers are flooded by end-users, how to efficiently manage virtual machines to balance both economical cost and ensure QoS becomes a mandatory work to service providers. Virtual machine migration feature brings a plenty of benefits to stakeholders such as cost, energy, performance, stability, availability. However, stakeholders' objectives are usually conflict with each other. Furthermore, the optimal resource allocation problem in cloud infrastructure is usually NP-Hard or NP-Complete class. In this paper, the virtual migration problem is formulated by applying the game theory to ensure both load balance and resource utilization. The virtual machine migration algorithm, named V2PQL, is proposed based on Markov decision process and Q-learning algorithm. The results of the simulation demonstrate the efficiency of our proposal which are divided into training phase and extraction phase. The proposed V2PQL algorithm has been benchmarked to the Round-Robin, inverse Ant System, Max–Min Ant System, and Ant System algorithms in order to highlight its strength and feasibility in extraction phase. [ABSTRACT FROM AUTHOR]

Cloud data center consumes enormous amount of electrical power resulting in high operating cost and carbon dioxide emission. Virtual machine (VM) consolidation strives to reallocate VMs to the minimum number of physical machines dynamically to reduce energy consumption. An approach is proposed to solve consolidation problem of VMs toward optimization of their energy consumption. A novel algorithm is presented, and an energy consumption model is also presented to support the algorithm to find optimal solutions. Experimental results show that energy consumption can be reduced greatly by dynamic VM consolidation. In comparison with other classic algorithms, our proposed algorithm outcomes them and achieves average savings of 42% on energy consumption. [ABSTRACT FROM AUTHOR]

Mobile edge computing (MEC) is an enabling technology for low-latency AI applications, by caching AI services originally deployed in remote data centers to 5G base stations in network edge. Due to limited computing resource of 5G base stations, not all services can be cached in base stations to meet the resource demands of user requests. Also, if the workload of a 5G base station reaches to its resource capacity, the energy consumption of the base station will be pushed up exponentially. To reduce the energy consumption and overcome resource limitations on base stations, an alternative is to allow the base stations to collaborate with each other to admit user requests. In this paper, we investigate the problem of collaborative service caching and request offloading between a 5G-enabled MEC and remote data centers, while meeting the quality of service (QoS) requirements of users, and resource capacities on base stations that are operated by multiple selfish network service providers. We aim to maximize the total payoff of all base stations. To this end, we first propose a two-stage optimization framework: In the first stage, we develop a mechanism that adopts a best-reply rule for dynamically distributed coalition formation. In the second stage, we propose a near-optimal payoff allocation method by devising a randomized algorithm with a provable approximation ratio. We then evaluate the performance of the proposed optimization framework by extensive experimental simulations. Simulation results show that the proposed framework outperforms its counterparts by achieving at least 30% higher payoff and 20% lower energy consumption of base stations. [ABSTRACT FROM AUTHOR]

With the development of the Internet of Things, the data that needs to be processed is increasing rapidly. Therefore, the collaboration of cloud and edge emerges as the times require. Edge nodes are mainly responsible for collecting data, and decide to process the data locally or offload to cloud data centers. Cloud data centers are suitable for data analysis, model training, and managing edge nodes. In this article, we focus on the task assignment problems in collaborative edge and cloud environments and study it in a distributed, non-cooperative environment. An M/M/1 queueing model is established to characterize the task transmission. Because of the multi-core processors, we set an M/M/C queueing model to characterize the task computation. We consider the problem from the perspective of game theory and formulate it into a non-cooperative game among multi-agents (multiple edge data centers) in which each agent is informed with incomplete information (allocation strategies) of others. For each agent, we define a function of the expected cost of tasks as the disutility function, and minimize it subject to the QoS constraint. We analyze the existence of Nash equilibrium and develop a Greedy Energy-aware Algorithm (GEA) to choose active servers using the Limit Searching Algorithm (LSA) to find the ceiling utilization. Then we propose the Best Response Algorithm (BRA) to optimize the utility function. The convergence of the BRA algorithm has been discussed. Finally, the results demonstrate that the BRA algorithm can get a solution close to Nash equilibrium and reach it quickly. [ABSTRACT FROM AUTHOR]

Future online services will be provisioned over the federation of infrastructure providers for economic reasons: for effective usage of resources and for a wide geographic reach of customers. The technical enablers of this setup, envisioned also for 5G services, are the virtualization techniques applied in data centers, and in access and core networks. Business aspects around the provisioning of these services however pose still unresolved questions. In this heterogeneous setup those who want to deploy an online service face the problem of selecting the compute and network resource set that fulfills the technical requirements of the service deployment and that is preferable also from an economic point of view. Infrastructure providers compete among themselves for these customers, and shape their business offerings with profit maximization in mind. We model this resource market with the tool set of graph and game theory in order to study its characteristics. We show that customers need to tackle a hard problem if they want to meet both their technical and business needs, let alone minimizing costs. Furthermore, we derive the best pricing strategies that the providers should follow given their expectation about customers' demand. Interestingly, we show that equilibrium prices and the attainable income strongly depend on the provider's location within the network. [ABSTRACT FROM AUTHOR]

Public buses have unique characteristics, such as fixed moving paths and time periods, uniform vehicular device standards, low risk of privacy exposure. This paper designed a public bus network-based vehicular crowd sensing system, considering the characteristics of public buses. In the system, data center utilized public buses of the bus network to collect urban data which was required by data users. It also studied the task assignment problem and the data trading problem in the system. This paper proposed an optimized task assignment strategy based on a greedy algorithm to minimize the system energy consumption of data collection, and proposed an optimal data trading strategy based on game theory to maximize the system utility. Finally, numerical results demonstrate the effectiveness of proposed strategies. [ABSTRACT FROM AUTHOR]

In this paper we propose an algorithm for load balancing in distributed data centers based on game theory. We model the load balancing problem as a non-cooperative game among the front-end proxy servers. We model the operating cost associated with a data center as a weighted linear combination of the energy cost and the latency cost. We propose a non-cooperative load balancing game with the objective of minimizing the operating cost and obtain the structure of Nash equilibrium. Based on this structure, a distributed load balancing algorithm is designed. We compare the performance of the proposed algorithm with the existing approaches. Numerical results demonstrate that the solution achieved by the proposed algorithm approximates the global optimal solution in terms of the cost and it also ensures fairness among the users. [ABSTRACT FROM AUTHOR]

The Nash Bargaining Solution (NBS) has been broadly suggested as an effective solution for the problem of fair allocation of multiple resources, namely bandwidth allocation in datacenters. In spite of being thoroughly studied, and provably strategy-proof for most scenarios, NBS-based allocation methods lack research on the strategic behavior of tenants in the case of proportionality of resource demands, which is common in datacenter workloads. We found that misbehavior is beneficial: by lying about bandwidth demands tenants can improve their allocations. We show that a sequence of selfish improvements leads to trivial demand vectors for all tenants. It essentially removes sharing incentives which are very important for datacenter networks. In this paper, we analytically prove that tenants can misbehave in 2- and 3- tenants cases. We show that misbehavior is possible in one recently proposed NBS-based allocation system if proportionality of demands is taken into account. Monte Carlo simulations were done for 2-15 tenants to show a misbehavior possibility and its impact on aggregated bandwidth. We propose to use another game-theoretic approach, namely Dominant Resource Fairness (DRF) to allocate bandwidth in the case of proportional demands. We show that this method performs significantly better than NBS after misbehavior. [ABSTRACT FROM AUTHOR]

The growth in computer and networking technologies over the past decades established cloud computing as a new paradigm in information technology. The cloud computing promises to deliver cost-effective services by running workloads in a large scale data center consisting of thousands of virtualized servers. The main challenge with a cloud platform is its unpredictable performance. A possible solution to this challenge could be load balancing mechanism that aims to distribute the workload across the servers of the data center effectively. In this paper, we present a distributed and scalable load balancing mechanism for cloud computing using game theory. The mechanism is self-organized and depends only on the local information for the load balancing. We proved that our mechanism converges and its inefficiency is bounded. Simulation results show that the generated placement of workload on servers provides an efficient, scalable, and reliable load balancing scheme for the cloud data center. Copyright © 2016 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Data centers hosting distributed computing systems consume huge amounts of electrical energy, contributing to high operational costs, whereas the utilization of data centers continues to be very low. Moreover, a data center generally consists of heterogeneous servers with different performance and energy. Failure to fully consider the heterogeneity of servers will lead to both sub-optimal energy saving and performance. In this study, we employ game theoretic approaches to model the problem of minimizing energy consumption as a Stackelberg game. In our model, the system monitor, who plays the role of the leader, can maximize profit by adjusting resource provisioning, whereas scheduler agents, who act as followers, can select resources to obtain optimal performance. In addition, we model the problem of minimizing average response time of tasks as a noncooperative game among decentralized scheduler agents as they compete with one another in the sharing resources. Several algorithms are presented to implement the game models. Simulation results demonstrate that the proposed technique has immense potential to improve energy efficiency under dynamic work scenarios without compromising service level agreements. [ABSTRACT FROM AUTHOR]

Demand response programs have been considered critical for power grid reliability and efficiency. Especially, the demand response of datacenters has recently received encouraging efforts due to huge demands and flexible power control knobs of datacenters. However, most current efforts focus on owner-operated datacenters, omitting another critical segment of datacenter business: multitenant colocation. In colocation datacenters, while there exist multiple tenants who manage their own servers, the colocation operator only provides facilities such as cooling, reliable power, and network connectivity. Therefore, colocation has a unique feature that challenges any attempts to design a demand response program: uncoordinated power management among tenants. To tackle this challenge, two incentive mechanisms are proposed to coordinate tenant power consumption for demand response under two different scenarios. First, in the case of economic demand response where the operator can adjust an elastic energy reduction target, we show that there is an interaction between the operator and tenant strategies, where each side maximizes its own benefit. Hence, we apply a two-stage Stackelberg game to analyze this scenario and derive this game’s equilibria. However, computing these equilibria can be intractable with exhaustive search; therefore, we propose an algorithm to find the Stackelberg equilibria with linear complexity. Second, in the case of emergency demand response where a fixed energy reduction target must be fulfilled, we devise two incentive schemes with the distributed algorithms that can achieve the same optimal social cost. While the first algorithm is based on the dual-decomposition method that is suitable for nonstrategic tenants, the second one is designed for strategic tenants to achieve a unique Nash equilibrium of a bidding game. Finally, trace-based simulations are also provided to illustrate the efficacy of our proposed incentive schemes. [ABSTRACT FROM PUBLISHER]

Data centers use dynamic virtual machine consolidation to reduce power consumption. Existing consolidation mechanisms are not efficient in cloud data centers, which have heterogeneous hardware infrastructure, huge scale, and highly variable non-stationary workloads. We use game theory to develop a novel distributed mechanism for both heterogeneous and homogeneous data centers of cloud computing. Our mathematical analysis shows that our mechanism converges after a finite number of migrations. In addition, we show that our worst case power consumption is only 23% more than the theoretical minimum. In order to validate our claim, we preform simulation in CloudSim with real workload traces from Google data centers. [ABSTRACT FROM AUTHOR]

Abstract: Cloud Gaming is a new kind of service, which combines the successful concepts of Cloud Computing and Online Gaming. It provides the entire game experience to the users remotely from a data center. The player is no longer dependent on a specific type or quality of gaming hardware, but is able to use common devices. The end device only needs a broadband internet connection and the ability to display High Definition (HD) video. While this may reduce hardware costs for users and increase the revenue for developers by leaving out the retail chain, it also raises new challenges for service quality in terms of bandwidth and latency for the underlying network. In this paper we present the results of a subjective user study we conducted into the user-perceived quality of experience (QoE) in Cloud Gaming. We design a measurement environment, that emulates this new type of service, define tests for users to assess the QoE, derive Key Influence Factors (KIF) and influences of content and perception from our results. [Copyright &y& Elsevier]

Summary: Nowadays, cloud computing has many benefits to accessibility, scalability, and cost‐effectiveness, leading to network security risks and vulnerabilities. Cloud computing is gaining in popularity with the advances and growth of its systems. Therefore, the security of this system and the identification of vulnerable data centers are more complicated than the past. Definitely discovering vulnerable data centers that are vulnerable to attacks can help to strengthen these data centers and provide a safer and more secure network structure. This paper examines the vulnerability of malware data centers in the infrastructure and cloud computing network structure. Based on the analysis of the cloud computing system in the field of game theory, we introduce a developed model for identifying vulnerable data centers in cloud computing. The developed model in this paper is based on the game theory as a mathematical tool. According to the game theory, we introduce a measure of the degree of vulnerability of data centers in the cloud computing network. [ABSTRACT FROM AUTHOR]