Search

With the growth of the cloud computing market, the number and scale of cloud data centers are expanding rapidly. While cloud data centers provide a large amount of computing power, generating tremendous energy consumption has become a fundamental issue in the financial and environmental fields. Improving quality of service and reducing energy costs are fundamental challenges for next-generation cloud data centers. Task scheduling in cloud data centers grows increasingly complex due to the heterogeneity of computing resources, intricate dependencies of jobs and rising expenses resulting from high energy consumption. Efficiently utilizing computing resources is crucial, so it is necessary to develop optimal strategies for job scheduling. This paper proposes a reinforcement learning-based task scheduler (E2DSched) for online scheduling of randomly arriving directed acyclic graph jobs in cloud data centers. E2DSched divides the scheduling process into three layers: task selection layer, server selection layer and frequency control layer. It achieves joint optimization of energy consumption and quality of service through three-layer cooperation. Finally, we compare E2DSched with various other algorithms, and the results show that E2DSched can provide excellent service with less energy consumption. [ABSTRACT FROM AUTHOR]

Virtualization technologies are extensively utilized in data centers, particularly cloud computing. This facilitates data center management and diminishes the number of physical machines (servers) and, subsequently, their cooling requirements, leading to cost, space, and power consumption reductions. When applications in data centers are executing independent parallel transactions, but with similar performance requirements, the appropriate level of virtual machine consolidation on a server poses a fundamental challenge for capacity planning. This article introduces a method to evaluate the performance speedup achieved through virtualization on any server and the effects of virtualization and consolidation overheads on physical or virtual machine scalability. This research formalizes the speedup and overheads, using classical computer architecture statements. but at the same time proposes a new method to analyze these overhead amounts and types, showing the scalability and efficiency of different consolidations in the same server and its comparison against no consolidation. This work also proposes a new way to determine the optimal number of physical servers and the optimal number of consolidated virtual machines for a given transaction workload. The real experimentation was performed with different workload sizes, types of virtualizations and different servers. The method presented also facilitates the representation of linear scalability against the real degree of parallelism of either physical machines or consolidated virtual machines for a given transaction workload, as well as striking the right balance between speedup and energy in virtual server consolidation. [ABSTRACT FROM AUTHOR]

While cloud computing is crucial in processing data from devices with low computational power, the latency introduced by the Internet backhaul limits real-time applications. By situating computing resources at the network's edge, edge computing offers low-latency services by offloading computations from high-performance computing (HPC) data centers to the edge servers, reducing wide Area network (WAN) strain. As a result, edge computing has unlocked opportunities for innovative applications that were previously unfeasible, such as connected vehicles or medical robotics. Nonetheless, deploying the infrastructure required to support edge computing services raises sustainability and energy consumption concerns. Consequently, the development of tools enabling researchers to explore innovative approaches to reducing the energy impact of edge computing is crucial. In this work, we present MintEDGE, a network simulator focused on the energy consumption of edge computing. Our simulator allows testing energy-saving approaches and task placement algorithms in realistic large-scale scenarios encompassing entire regions. [ABSTRACT FROM AUTHOR]

A workflow is sent from the internet of things (IoT) in the real world to a distributed edge micro data centre (EMDC), which is assigned to various resources for the necessary computation process. Since optimising energy consumption and other computing processes are crucial, a significant issue arises involving energy consumption and task execution time in an EMDC-integrated computing system. This study presented a task scheduling and allocation method with two phases in optimising the energy consumption and execution time of scientific workflow processing. The first phase assigned the workflow as a Directed Acyclic Graph (DAG) to an EMDC with the least latency. A selection algorithm then considered location routing problems (LRP), EMDC queue waiting time to initiate scheduling a DAG task and the EMDC capacity. The second phase demonstrated the allocation and schedule of DAG tasks to the resources of selected EMDCs. Subsequently, a critical path (CP) was proposed using the Multi-Objective Firefly Optimisation with Dynamic Voltage and Frequency Scaling (CP-MOFFO-DVFS) algorithm. This study demonstrated a 20 to 37% improvement in energy savings than the state-of-the-art method, thus substantially reducing execution time. [ABSTRACT FROM AUTHOR]

With the expansion of network scale, it is essential to study network reliability through connectivity and diagnosability. For Hypercube, DCell, BCube, and other networks, their connection modes are specific and different, so it is necessary to use different methods to study the properties of networks. This paper proposes a class of topological structure—cycle composition networks (CCNs), which not only contains k-ary n-cube and BC graph, but also includes the data center network CamCube and many other unknown networks. We then study their diameters and path construction algorithm in them. Furthermore, we establish their classical connectivity and diagnosability under the PMC and the MM ∗ models, respectively. Finally, we give the 1-good-neighbor connectivity and diagnosability of the CCNs under the PMC model for n ⩾ 3 and l = 2 . [ABSTRACT FROM AUTHOR]

Recently, a growing amount of time series data has been collected in cloud data centers, making anomaly detection for multivariate time series analysis increasingly necessary. However, extracting meaningful features from multivariate time series remains challenging due to the limited amount of labeled data and highly complex temporal correlations. Additionally, many unsupervised deep learning methods often result in a high false alarm rate. This study proposes a new unsupervised multivariate time series anomaly prediction model called the Predictive Wasserstein Generative Adversarial Network with Gradient Penalty (PW-GAN-GP). Our model adopts both Wasserstein Distance and Gradient Penalty, making the adversarial training more stable and helping the generator's output to more closely resemble the real data. Moreover, a novel anomaly score function combining reconstruction, discrimination, and prediction errors is used to improve precision while maintaining recall. The experimental results on four public cloud computing datasets demonstrate that our proposed PW-GAN-GP outperforms the suboptimal baseline, with improvements of 22.11% and 13.47% in precision and F1 scores, respectively. [ABSTRACT FROM AUTHOR]

Decreasing carbon emissions of data centers while guaranteeing Quality of Service (QoS) is one of the major challenges for efficient resource management of large-scale cloud infrastructures and societal sustainability. Previous works in the area of carbon reduction mostly focus on decreasing overall energy consumption, replacing energy sources with renewable ones, and migrating workloads to locations where lower emissions are expected. These measures do not consider the energy mix of the power used for the data center. In other words, all KWh of energy are considered the same from the point of view of emissions, which is rarely the case in practice. In this paper, we overcome this deficit by proposing a novel practical CO2-aware workload scheduling algorithm implemented in the Kubernetes orchestrator to shift non-critical jobs in time. The proposed algorithm predicts future CO2 emissions by using historical data of energy generation, selects time-shiftable jobs, and creates job schedules utilizing greedy sub-optimal CO2 decisions. The proposed algorithm is implemented using Kubernetes' scheduler extender solution due to its ease of deployment with little overheads. The algorithm is evaluated with real-world workload traces and compared to the default Kubernetes scheduling implementation on several actual scenarios. [ABSTRACT FROM AUTHOR]

This study addresses the energy efficiency challenge in cloud data centers by examining the Virtual Machine Placement (VMP) problem. VMP involves mapping virtual machines (VMs) to physical machines (PMs) under capacity constraints. The paper focuses on the bin packing with linear usage cost (BPLUC) variant of bin packing, which includes fixed and variable costs in the calculation of the cost of a used bin. We prove that every approximation algorithm for the bin and vector bin packing can be used for BPLUC and VBPLUC, respectively. We propose a more power-efficient approach to VMP by applying a vector bin packing algorithm to minimize power consumption in data centers. We test the proposed algorithm on various synthetic and real workloads, and the experimental results demonstrate that it is more power-efficient than existing algorithms for VMP. The findings suggest that the proposed algorithm has significant implications for energy-efficient strategies in cloud data centers. Generally, this study makes contributes to the development of energy-efficient approaches to VMP that can help reduce power consumption and improve the sustainability of cloud data centers. [ABSTRACT FROM AUTHOR]

The cloud-computing paradigm based on containers has progressively grown in recent years as a flexible strategy that has proven to be energy efficient. The increasing usage of the container as a service technology in data centers (DCs) among cloud providers highlights the necessity of the container installation design phase in cloud environments. Cloud providers attempt to enhance resource utilization and reduce energy consumption by employing various VM selection and placement policies. This procedure for placement acquires a new aspect, with containers now being deployed on virtual machines (VMs) and those guest VMs being installed on physical machines (PMs). The intricacy of this issue increases when the variety of the containers, VMs, and PMs is taken into account. In this paper, an optimal placement strategy for containers is proposed based on the bio-inspired algorithms. The firefly algorithm has been modified to use discretization strategy (Discrete Firefly Algorithm, DFF) and has also used local search mechanism (Discrete Firefly with Local Search Mechanism, DFFLSM). The proposed versions of firefly algorithm are compared with first fit, first fit decreasing, random algorithm and ant colony algorithm. The comparison is done based on average energy consumption, average active VM, average active PM and average overall service-level agreement violations in the DC. The results show that DFFLSM performs better than all pre-existing container placement algorithms in terms of energy efficiency. It reduces average energy consumption of DC by 9.32% and 40.85% and average active PM by 18.30% and 21.89% in homogenous and heterogeneous environment, respectively. [ABSTRACT FROM AUTHOR]

Over the years, data centers have been evolving to meet the new demands of cloud computing platforms, e-commerce, social networking services, and big data. These large data centers must meet various dependability requirements to guarantee the quality of service at a high level of reliability and availability, reducing interoperability time, as this is a major competitive factor for companies. The cooling system plays an important role in the availability of data centers. It is responsible for keeping the IT system at a suitable temperature to avoid hardware and software failures since operating at high temperatures negatively impacts the reliability of the electronic components. This paper investigates reliability and availability metrics and performs parametric sensitivity analysis for data center cooling systems. Our approach is based on modeling reliability block diagrams and the sensitivity analysis technique; the last is to assess how sensitive the availability is about model component failure and repair times. [ABSTRACT FROM AUTHOR]

The design-building of IT services in data centers has been historically conducted by applying rigorous IT service design-building phases-workflows. Consequently, relevant research has been conducted upon these rigorous phases-workflows to provide theoretical foundations and practical guidance to IT service design-building architects. However, the current dynamic business-governmental environment is demanding agile approaches, and research from this perspective is still very scarce. This research, thus, aims to provide an updated review and quantitative evaluation of the main rigorous (ITIL v2011, CMMI-SVC v1.3, and ISO/IEC 20000-1:2018) and main agile (ITIL v4, VeriSM, and ISO/IEC 29110-4-3) IT service design-building phases-workflows. For this aim, we conduct a conceptual review research methodology enhanced with an analytics hierarchical process (AHP) method to assess quantitatively how well these six IT service design-building phases-workflows fit two theoretical expected rigorous and agile IT service design-building phases-workflows pro formas. We found that the ITIL v2011 and CMMI-SVC v1.3 phases-workflows fit the rigorous pro forma with a high level, and the ISO/IEC 20000-1:2018 fits a moderate level, whereas all these three ones fit a low level the agile pro forma, as it was expected. ITIL v4 and VeriSM were found to fit a high level the agile pro forma but the ISO/IEC 29110-4-3 fits a moderate level. Unexpectedly, ITIL v4 and the ISO/IEC 29110-4-3 fit a moderate and moderate levels the rigorous pro forma, but VeriSM fits a low level as it was expected. Hence, we can conclude that ITIL v4 and VeriSM provide the most agile IT service design-building phase-workflow, and CMMI-SVC v1.3 the most rigorous one. The ISO/IEC 20000-1:2018 and ISO/IEC 29110-4-3 standards are still aligned to the rigorous approach, and ITIL v4 exhibits a dual moderate rigorous and high agile profile. Hence, this research provides ITSM professionals with an updated analysis useful to guide the selection and application of the IT service design-building phase-workflow—rigorous or agile one—in data centers. This research also contributes to the IT service design-building literature with updated insights and proposes specific research avenues to advance our scientific knowledge on how to design-building IT services. [ABSTRACT FROM AUTHOR]

The systematic design, building and implementation of IT services for on-premise, cloud-based or hybrid data centers are a hard engineering and managerial process. Heavyweight IT service design workflows—from ITIL v2011, CMMI-SVC v1.3 and the ISO/IEC 20000 standard—have been used in the last decade. However, the current dynamic business requirements demand to accelerate IT service design, building and implementation. Consequently, agile workflows—from ITIL v4 and VeriSM—or lightweight ones—from ISO/IEC 29110-4-3 standard and FitSM—have been proposed. However, despite these IT service design workflows provide high valuable content, they are still reported in a coarse-grain level, and thus academics and practitioners interested in using to them must add ad-hoc roles, activities-tasks, and artifacts-templates. In this research, we address this relevant knowledge gap, and using a Design Science Research Methodology (DSRM), we elaborate and evaluate exploratory an agile IT service design workflow—called AVS_FD_MVITS—that provides fine-grain content regarding roles, activities-tasks and artifacts-templates. AVS_FD_MVITS was designed using design components taken from Scrum-XP and ITIL v4 IT service design workflows. To evaluate exploratory AVS_FD_MVITS, we applied it in a realistic case in an academic small data center. The exploratory qualitative-based evaluation suggests that AVS_FD_MVITS can be considered a valid agile IT service design workflow useful for academics and practitioners. However, confirmatory quantitative-based evaluations are asked for advancing this research stream. [ABSTRACT FROM AUTHOR]

Container-based Hadoop distributed file system (HDFS) storage has been widely used in cloud data center networks, while traditional HDFS has single point problem resulting in overall unavailability. In this paper, we mainly study the storage reliability of the Docker container-based HDFS cluster with single point of failure. Firstly, we investigate a data volume-based persistence solution of Hadoop with the single point failure and single backup strategy of HDFS cluster. Secondly, we propose an HDFS-based replica placement algorithm for data storage with considering the performance of the host and container nodes. Thirdly, we design the KADC-KNN data segmentation algorithm to effectively store the persistent data of the Docker container. Extensive experimental results show that this method can effectively ensure the stable storage and fast migration of cluster data. Compared with the most advanced algorithm, the proposed data volume persistence algorithm DVPS can improve the data reliability by 19.8%. The data partitioning algorithm KADC-KNN improves the partitioning accuracy by 20.2% and has lower time overhead. [ABSTRACT FROM AUTHOR]

Every computation presents a small chance that an unexpected phenomenon ruins or modifies its output. Computers are prone to errors that, although may be very unlikely, are hard, expensive or simply impossible to avoid. In the exascale, with thousands of processors involved in a single computation, those errors are especially harmful because they can corrupt or distort the results, wasting human and material resources. In the present work, we study the effect of ionizing radiation on several pieces of commercial hardware, very common in modern supercomputers. Aiming to reproduce the natural radiation that could arise, CPUs (Xeon, EPYC) and GPUs (A100, V100, T4) are subject to a known flux of neutrons coming from two radioactive sources, namely 252 Cf and 241 Am-Be, in a special irradiation facility. The working hardware is irradiated under supervision to quantify any appearing error. Once the hardware response is characterised, we are able to scale down the radiation intensity and to estimate the effects on standard data centres. This can help administrators and researchers to develop their contingency plans and protocols. [ABSTRACT FROM AUTHOR]

Generally, seismic data have discontinuity or planar fracture anomalies in the volume of rocks, commonly known as seismic faults. A large number of methods exist to interpret seismic faults, nevertheless automating the process prevails to be a practical challenge. We propose a graph representation-based approach for interpreting faults in seismic data using the graph convolutional network (GCN). We extract 2D patches of data centered around seismic data points from the 3D seismic volumes for training. Then we represent these patches in the graph domain using the k-nearest neighbor graphs followed by the application of GCN. After training the patches in the networks, we classify the patches to identify the faults. We consider both synthetic and real data for training and testing. The seismic amplitude values, the seismic attribute values, and the successive difference values are used in the networks. We compare our implementation with the state-of-the-art method, the convolutional neural networks (CNN). The results show good accuracy when applied to both synthetic and real data. Additionally, the GCN method is more time efficient than that of CNN. We process the 3D seismic section by parallel processing the 2D patches. [ABSTRACT FROM AUTHOR]

VMP (Virtual Machine Placement) is a crucial technology for energy consumption optimization of the cloud data center. Existing works mainly focus on virtual machine consolidation to increase resource utilization and reduce computing energy consumption. However, existing studies usually ignore the thermal effect that an intensive workload on IT (Information Technology) equipment can raise energy consumption by cooling systems and generate hotspots. In addition, an excessive number of virtual machine migrations increases migration costs and risks violating the SLA (Service Level Agreement) signed with users. In this paper, we present a comprehensive system model and formulate the problem as a constrained multi-objective optimization. We propose a novel thermal-aware VMP strategy to solve the problem by jointly considering virtual machines' migration cost, energy consumption, and heat recirculation around server racks. Our strategy makes placement decisions using MOPFGA (Multi-objective algorithm based on Pathfinder Algorithm and Genetic Algorithm) that combines classic MOPFA and GA enhanced by OBL (Opposition Based Learning) for fast convergence and avoidance of local optimum. Extensive experiments based on CloudSim using real data center workload data from PlanetLab show that our algorithm overcomes the defects of the MOPFA (multi-objective pathfinder algorithm) and GA (genetic algorithm) and significantly improves the overall efficiency of a data center. Compared with several state-of-the-art algorithms, MOPFGA on average reduces virtual machine migrations by 77.52%, increases CRAC (Computer Room Air Conditioner) supply temperature by 1.24%, and reduces cooling energy consumption by 24.78% and computational energy consumption by 23.62%. [ABSTRACT FROM AUTHOR]

As the volume of data keeps growing rapidly, more and more storage devices, servers, and network devices are continuously added into data center networks (DCNs) to store, manage, and analyze the data. The industry experience indicates that, instead of adding a huge number of servers into the DCNs at a time, the DCN can also be expanded gradually by adding a small number of servers from time to time. This paper proposes a new type of dual-centric data center network structure, called GHDC (Generalized Hypercube Data Center network architecture), which is constructed by using commodity switches and multi-port servers. After analyzing the shortest distance between any two vertices, a routing algorithm for GHDC is developed. To achieve incremental scalability, two incomplete GHDC structures are proposed. A small number of servers can be added into the incomplete GHDC structures while their topological properties are maintained. The analysis and experiment results show that GHDC significantly outperform the other DCN structures, such as FatTree, BCube, Platonica, FCell, FSquare, and FRectangle, in terms of the incremental scalability and robustness. The average throughput of GHDC is approximately comparable to that of FatTree, BCube, and FSquare, and is higher than that of Platonica, FCell, and FRectangle by about 130.2%, 17.45% and 25.5%. Compared with the FatTree, BCube, FCell, FRectangle, and FSquare, GHDC reduces the cost by about 68.49%, 78.04%, 10.84%, 22.85%, and 29.55%, and reduces the max energy consumption by about 69.45%, 34.48%, 11.55%, 24.31%, and 29.58%, respectively. The actual energy consumption of GHDC is much lower than that of FatTree, BCube, FCell, FRectangle, and FSquare, and is little higher than that of Platonia. [ABSTRACT FROM AUTHOR]

At present, most causal consistency models based on cloud storage can no longer meet the needs of delay-sensitive applications. Moreover, the overhead of data synchronization between replicas is too high. This paper proposes a causal consistency model of edge–cloud collaborative based on grouping protocol. The model is based on the edge–cloud collaboration architecture, partitions cloud data centers and groups edge nodes by distributed hash tables, and stores a subset of the complete data set in nodes located at the edge of the network, thereby realizing partial geo-replication in edge–cloud collaboration environment. At the same time, we design a group synchronization algorithm called Imp_Paxos, so that the update only needs to be synchronized to the main group, which reduces the visibility delay of the update and decreases the data synchronization overhead. Besides, a sort timestamp is proposed in this paper, which generates different timestamps according to the type of update to track causality, keeping the amount of metadata managed in a relatively stable and low state. Therefore, the proposed model reduces the overhead of metadata for system management and improves throughput quantity of system. Experiments show that our model performs well in terms of throughput, operation latency, and update visibility latency compared with existing causal consistency models. [ABSTRACT FROM AUTHOR]

Software-defined networking (SDN) separates the forwarding plane and control plane of the network equipment, adopts a centralized control mode to simplify network deployment and improve network management efficiency and realizes the network flexible control and management of traffic through programmable open interfaces. At present, it has been widely used in domestic and international data centre networks. With the explosive growth of the scale of data centres and the increase in user requirements for service quality, load balancing and congestion control of data centres have become significant issues in current research. After studying and analysing data centre SDN architecture and load balancing problems in detail, certain experts proposed an SDN load balancing algorithm, based on improved ant colony optimization-load balancing (IACO-LB). Firstly, the overall framework of SDN load balancing in data centres is studied, which is mainly divided into three parts: basic network equipment, OpenFlow protocol and controller. Among them, the controller constitutes the core of the entire load balancing system, including four modules: network topology awareness, status collection, the core of the load balancing algorithm and the flow table distribution. Then, an SDN load balancing algorithm, based on the improved ant colony optimization (IACO) is proposed to achieve dynamic load balancing of the SDN. The algorithm fully considers the performance parameters of network links and servers, and its design is based on the principle of selecting links and servers with low utilization. The evaluation methods of server module and link module are designed and the ant colony algorithm is used to find the global optimal solution. In order to prevent the algorithm from falling into local optimum, the Kent chaotic model is adopted to disturb the transition probability of the ant colony, by improving the basic ant colony algorithm. Finally, a network topology model was established in MATLAB to carry out simulation experiments. The results show that, compared with the equivalent multi-path algorithm and path server traffic scheduling algorithm, IACO-LB can effectively solve the load balancing problem of SDN and can dynamically adjust the routing scheme, according to the changes in network link traffic and server utilization. The algorithm converges quickly and can achieve a better global load balancing scheme. [ABSTRACT FROM AUTHOR]

Explicit Congestion Notification (ECN) is designed for single queues. However, today, data center networks (DCNs) need multiple queues on each switch port. But, if some of the switches in multiple queue scenarios exceed the ECN marking threshold, all packets on the same port can receive the ECN mark. To solve this problem, we propose mapping-ECN as a systematic answer to the wrong marking problem. First, we differentiate the mice and elephant flows learning algorithm. Then, we prioritize mice flows by keeping in mind the deadline of other flows to not sacrifice them. Secondly, if a packet is marked, we need to have the privilege of using a faster path than other packets for early notification of network status. This will give a complete picture of the instant requests from all senders. In the worst case, if there is no capacity in the buffer to transmit the packets that exceed the threshold of the buffer, mapping-ECN uses Cut Payload (CP), where CP drops the payloads of the packets when a queue reaches the threshold, rather than the metadata. Consequently, just one bit will transmit that carries the information of the packet. Therefore, the sender will immediately retransmit that packet without waiting for a time-out like TCP. This retransmission can arrive within a millisecond for having an extremely low latency network. Last but not least, mapping-ECN explores different kinds of neural network techniques to avoid miss marking in the output port buffer. Therefore, if any packet is marked within the queue buffer, these marked packets are not considered again for marking choices within the output port buffer. Mapping-ECN improves the overall performance of Flow-Completion Time (FCT) for short flows around 7%, 99th percentile around 52%, and FCT for short flows around 8% in comparison between MQ-ECN. Moreover, when compared to the MQ-ECN, Mapping-ECN improves the FCT for large flows, for cache flows and for mice (web search) flows 4, 15 and 6%, respectively. This improvement is legible in comparison between DemePro and Priority-ECN as well. [ABSTRACT FROM AUTHOR]

Real-time tracking and surveillance of patients' health has become ubiquitous in the healthcare sector as a result of the development of fog, cloud computing, and Internet of Things (IoT) technologies. Medical IoT (MIoT) equipment often transfers health data to a pharmaceutical data center, where it is saved, evaluated, and made available to relevant stakeholders or users. Fog layers have been utilized to increase the scalability and flexibility of IoT-based healthcare services, by providing quick response times and low latency. Our proposed solution focuses on an electronic healthcare system that manages both critical and non-critical patients simultaneously. Fog layer is distributed into two halves: critical fog cluster and non-critical fog cluster. Critical patients are handled at critical fog clusters for quick response, while non-critical patients are handled using blockchain technology at non-critical fog cluster, which protects the privacy of patient health records. The suggested solution requires little modification to the current IoT ecosystem while decrease the response time for critical messages and offloading the cloud infrastructure. Reduced storage requirements for cloud data centers benefit users in addition to saving money on construction and operating expenses. In addition, we examined the proposed work for recall, accuracy, precision, and F-score. The results show that the suggested approach is successful in protecting privacy while retaining standard network settings. Moreover, suggested system and benchmark are evaluated in terms of system response time, drop rate, throughput, fog, and cloud utilization. Evaluated results clearly indicate the performance of proposed system is better than benchmark. [ABSTRACT FROM AUTHOR]

The resources required to service cloud computing applications are dynamic and fluctuate over time in response to variations in the volume of incoming requests. Proactive autoscaling techniques attempt to predict future resource demand and assign resources before the request. Thus, establishing accurate prediction methods optimizes resource allocation and avoids service-level agreement violations (SLA). Most existing methods for predicting resource usage in cloud computing rely on univariate time series prediction models. These models consider only a single resource usage metric, make future predictions based on historical data for the target metric, and provide predictions for a single step ahead. This paper provides a hybrid method for multivariate time series workload prediction of host machines in cloud data centers that predicts the workload for the next few steps. First, a statistical analysis is used to construct the training set. Then, a convolutional neural network (CNN) is employed to extract the hidden spatial features between all correlated variables. Finally, the spatial features extracted by the CNN are fed into a GRU network optimized with the attention mechanism in order to extract the temporal correlation features. Two experiments employing Google cluster data were conducted to evaluate the proposed method. Experimental results reveal that our method improves prediction accuracy by 2% to 28% compared to baseline methods and previous research. [ABSTRACT FROM AUTHOR]

How to improve resource utilization of cloud data centers (CDCs) and ensure users' quality of service (QoS) through efficient virtual machine (VM) scheduling is an urgent problem. Especially when service reliability is taken into consideration, the problem becomes more challenging. However, existing related researches mostly ignore the influence of reliability factors, such as failures and recoveries of computing nodes (CNs), which cannot reflect the realistic situations of real-life CDCs. Therefore, this paper investigates the problem of fault tolerance-aware VM scheduling and formulates it as a multi-objective optimization model with multiple QoS constraints. The proposed model tries to minimize users' total expenditure and, at the same time, maximize the successful execution rate of their businesses. To solve the proposed optimization model, a greedy-based best fit decreasing (GBFD) algorithm is then developed. The GBFD algorithm adopts a cost efficiency factor whose definition is according to the characteristics of CNs, to select a suitable CN for each VM request. Finally, extensive experiments are conducted to verify the feasibility of the proposed models and algorithm based on both the real-world CDC cluster data sets and the simulation ones. The results show that, first, as expected, fault tolerance significantly influences the performance criteria of VM scheduling and second, in most cases, the developed algorithm can decrease users' expenditure, increase success rate for executing their business and improve their overall satisfactions. Specifically, under real-world CDC cluster scenario, GBFD algorithm can increase the overall satisfaction of all cloud users by 38.3%, 20.9% and 14.6%, respectively, compared with the other three ones. Thus, the developed algorithm can perform better under fault tolerance-aware cloud environments. [ABSTRACT FROM AUTHOR]

Data center network plays an important role in improving the performance of cloud computing. Hamiltonian properties and Hamiltonian connectivity have important applications in communication network. The existence of Hamiltonian path can make the network more efficient communication. HCN and BCN networks are two important data center networks with nice routing performance and excellent scalability. In this paper, we study the Hamiltonian properties and disjoint path covers of these two networks. Firstly, we prove that HCN(n, h) is Hamiltonian-connected with n ≥ 4 and h ≥ 0 . Secondly, we prove that BCN (α , β , h , γ) is Hamiltonian-connected with h < γ , α ≥ 4 , β ≥ 1 , h ≥ 0 , γ ≥ 0 . Finally, we design Hamiltonian path construction algorithms for HCN and BCN networks. Simulation experiments verify the construction process of Hamiltonian path. Moreover, the running time of the routing algorithm designed in this study is compared with the classical shortest path multicast tree algorithm DijkstraSPT, and its running time is lower than that of the algorithm DijkstraSPT by about 5ms on different server nodes, which shows that the routing algorithm designed in this study according to HCN and BCN structure operate efficiently. [ABSTRACT FROM AUTHOR]

Resource optimization is becoming a prime factor in the progress of Internet-based technology, Cloud Computing. A resource management model is highly required in cloud data center paradigms to utilize available resources effectively. Bin-Packing problem is an applicable combinatorial optimization for Virtual Machine (VM) to Physical Machine (PM) allocation to minimize the required PMs. In this paper, we have proposed an efficient resource allocation and management algorithm in two phases. During the first phase, a Load Balanced Multi-Dimensional Bin-Packing (LBMBP) heuristic for Virtual Machine (VM) to Physical Machine (PMs or host) allocation is introduced, considering multidimensional resources: CPU, RAM, and Network Bandwidth. In the Second Phase, to perform VM migration, a mechanism to detect overloaded and underloaded hosts based on outliers has been described. The proposed work illustrated the simulation results using CloudSim Plus Simulator and observed a reduction in the number of active PMs. Energy consumption and the number of migrations with improved resource utilization. [ABSTRACT FROM AUTHOR]

Data centers are a fundamental infrastructure in the Big-Data era, where applications and services demand a high amount of data and minimum response times. The interconnection network is an essential subsystem in the data center, as it must guarantee high communication bandwidth and low latency to the communication operations of applications, otherwise becoming the system bottleneck. Simulation is widely used to model the network functionality and to evaluate its performance under specific workloads. Apart from the network modeling, it is essential to characterize the end-nodes communication pattern, which will help identify bottlenecks and flaws in the network architecture. In previous works, we proposed the VEF traces framework: a set of tools to capture communication traffic of MPI-based applications and generate traffic traces used to feed network simulator tools. In this paper, we extend the VEF traces framework with new communication workloads such as deep-learning training applications and online data-intensive workloads. [ABSTRACT FROM AUTHOR]

The dynamic nature of the cloud environment increases the complexity of managing its resources and the distribution of user workload between the available containers in the data center. However, the workload must be balanced to improve the cloud system's overall performance. Generally, most of the existing load balancing techniques suffer from performance degradation due to the communication overheads among the containers. Moreover, less attention is given to stabilize the load in a multi-cloud environment. Therefore, to overcome this problem, there is a need to develop an elastic load balancing method to improve the performance of cloud systems. This paper proposed an autonomic CSO-ILB load balancer to ensure the elasticity of the cloud system and balance the user workload among the available containers in a multi-cloud environment. The concept of multi-loop has been utilized in our approach to enabling efficient self-management before load balancing. The tasks are scheduled to the containers using an extended scheduling algorithm called Deadline-Constrained Make-span Minimization for Multi-Task Scheduling (DCMM-MTS). Based on the task scheduling, the load in each container is computed and then balanced using the proposed load balancer algorithm CSO-ILB. The proposed approach is evaluated in the Container CloudSim platform, and the performance is compared with the existing meta-heuristic algorithms such as Ant Colony Optimization, Bee Colony Optimization, Shuffled Frog Leaping Algorithm and Cat Swarm Optimization (CSO). The simulations proved that the proposed approach outperformed the other approaches in terms of reliability, CPU utilization, make-span, energy utilization, response time, execution cost, idle time, and task migration. [ABSTRACT FROM AUTHOR]

Resource provisioning for cloud computing requires adaptive and accurate prediction of cloud workloads. However, existing studies in workload prediction have faced significant challenges in predicting time-varying cloud workloads of diverse trends and patterns, and the lack of accurate prediction often results in resource waste and violation of Service-Level Agreements (SLAs). We propose a bagging-like ensemble framework for cloud workload prediction with Adaptive Pattern Mining (APM). Within this framework, we first design a two-step method with various models to simultaneously capture the "low frequency" and "high frequency" characteristics of highly variable workloads. For a given workload, we further develop an error-based weights aggregation method to integrate the prediction results from multiple pattern-specific models into a final result to predict a future workload. We conduct experiments to demonstrate the efficacy of APM in workload prediction with various prediction lengths using two real-world workload traces from Google and Alibaba cloud data centers, which are of different types. Extensive experimental results show that APM achieves above 19.62% improvement over several classic and state-of-the-art workload prediction methods for highly variable real-world cloud workloads. [ABSTRACT FROM AUTHOR]

Due to the increasing use of sensors/devices in smart cities, IoT/cloud data centers must provide adequate computing resources. Efficient resource management is of the biggest challenges in distributed computing. This research proposes a solution to use the activity log of sensors to extract their activity patterns. These patterns contribute to the resource management to predict future resource requirements and act accordingly. In this framework, called sensor mining at resource management framework, the pattern extraction can be performed by the sensor mining algorithm at cloud-only data centers (SM@RMFCLOUD) or cloud/fog servers (SM@RMFFOG). Experiments apply both CityPulse and ideal datasets to evaluate the presented frameworks. The sensor mining in both cloud/fog and cloud-only frameworks improves throughput, response time, and execution delay without increasing costs, energy, and bandwidth consumption compared to the frameworks with no sensor mining. [ABSTRACT FROM AUTHOR]

Data consistency has always been a significant topic in distributed systems. In the existing consistency models, causal consistency attracts more attention because it can meet high-performance requirements even when there are network partitions in the system. The synchronization method between replicas is one of the key indicators affecting the performance of causal consistency, especially when there are a large number of nodes in the system. In the case of deploying a large number of nodes in the system, this paper optimizes the synchronization mode between data centers and proposes a causal consistency model based on the grouping strategy (Gart). Gart manages all nodes in groups to reduce the management cost during data synchronization, and adopt a leader mechanism to improve the management efficiency of the system. At the same time, a client migration mechanism be introduced to ensure that throughput can be improved without sacrificing the remote update visibility. The simulation results demonstrate that compared with the existing causal consistency model, Gart can achieve better throughput when handling a large number of nodes, and with the same communication delay, it can achieve higher update visibility. [ABSTRACT FROM AUTHOR]

In recent years, data centers (DCs) have evolved a lot, and this change is related to the advent of cloud computing, e-commerce, services aimed at social networks, and big data. Such architectures demand high availability, reliability, and performance at satisfactory service levels; requirements are often neglected at the expense of high costs. In addition, the use of techniques capable of promoting greater environmental sustainability is most often forgotten in the design phase of such architectures. Approaches to perform an integrated assessment of dependability attributes for DCs, in general, are not trivial. Thus, this work presents the dependability attributes (availability and reliability), performability, and sustainability parameters that need special attention in implementing a cooling subsystem in DCs. That is one of the most cost generators for these infrastructures. In this study, we use the hypothetical-deductive method through a quantitative and qualitative approach; as for the procedure, it is bibliographical research through the review of scientific studies, and the research objectives are exploratory in nature. The results show that among all the papers selected and analyzed in this systematic literature review (SLR), none have jointly addressed performability, dependability, and sustainability in cooling systems for DCs. The main results of this work are presented through research questions, as they bring evidence of gaps to be addressed in the area. The four research questions point out challenges in implementing cooling systems in DCs and present the techniques and/or methods most used to propose or analyze data center cooling infrastructures; addressing the essential sustainability requirements for cooling subsystems, and finally, presenting open questions that can be investigated in the area of sustainable cooling in DCs regarding the data center's cooling and the difficulty of incorporating dependability attributes in the environmental context. In addition to these results, the present study actively contributes to the concept of a "green data center" for the companies, which ranges from the choice of renewable energy sources to more efficient information technology equipment. Hence, we show the relevance and originality of this SLR and its results. [ABSTRACT FROM AUTHOR]

Thermal-aware (TA) task allocation is one of the most effective software-based dynamic thermal management techniques to minimize energy consumption in data centers (DCs). Compared to its counterparts, TA scheduling attains significant gains in energy consumption. However, the existing literature overlooks the heterogeneity of computing elements in terms of thermal constraints while allocating or migrating user jobs, which may significantly affect the reliability of racks and all the equipment therein. Moreover, the workload distribution among these racks/servers is not fair and efficient in terms of thermal footprints; it is potentially beneficial to determine the workload proportion for each computing node (rack/server) based on its marginal contribution in disturbing the thermal uniformity (TU) in a DC environment. To solve the said problems, we model the workload distribution in DCs as a coalition formation game with the Shapley Value (SV) solution concept. Also, we devise Shapley Workload (SW), a TA scheduling scheme based on the SV to optimize the TU and minimize the cooling cost of DCs. Specifically, the scheduling decisions are based on the ambient effect of the neighboring nodes, for the ambient temperature is affected by the following two factors: (1) the current temperature of computing components and (2) the physical organization of computing elements. This results in lower temperature values and better TU, consequently leading to lower cooling costs. Simulation results demonstrate that the proposed strategy greatly reduces the total energy consumption compared to the existing state-of-the-art. [ABSTRACT FROM AUTHOR]

A Data Center Network (DCN) is composed of a large number of computing and storage nodes that are interconnected by well-organized switches. The Software-Defined Networking (SDN) based DCN (SD-DCN) improves resource utilization and provides virtual network access by separating the data plane and control plane of DCN. However, the routing strategies in current SD-DCN systems are based on traditional mechanisms that lack in real-time modification and are less efficient in resource utilization. To overcome these limitations, Convolutional Neural Network (CNN) deep learning model is proposed in this paper to improve the routing computation in SD-DCN, i.e., FAT-tree topology. The CNN deep learning model gives intelligent paths according to online training of traffic patterns. Moreover, the achieved network performance is compared with specific existing routing algorithms for SD-DCN. It is observed that the average network throughput is almost doubled for hot-spot traffic as compared with existing routing algorithms OSPF and FlowDCN. The experimental results show that, compared to ANN, the proposed model has increased the average network throughput by approximately 40%. Also, the proposed CNN model has outperformed the Artificial Neural Network (ANN) model in terms of average network delay and packet loss rate. Similarly, the overall bandwidth utilization is achieved by approximately 70% as compared to existing mechanisms. [ABSTRACT FROM AUTHOR]

New hierarchical crossbar switch architectures, such as Omni-Path (OPA) and Cray X2, have appeared to improve packet latency, reduce overall cost and increase fault tolerance of the high-performance interconnection networks in supercomputing and data center systems. These and other interconnect technologies (Infiniband or 40/100 Gigabit Ethernet) include support to provide quality of service (QoS) to the applications. In this paper, we show how this QoS support can be enabled to achieve bandwidth and/or latency differentiation in Omni-Path interconnection networks, as a representative case of hierarchical switches. To do that, three different table-based schedulers are used. We include the description of these schedulers and a comparative study by using the results obtained when we evaluate them with Hiperion, a simulation tool that implements an OPA model. [ABSTRACT FROM AUTHOR]

Complex resource usage patterns of scaling Cloud workloads and heterogeneous infrastructure remain a challenge for accurate modelling of server load, which is the key to effective capacity sizing and provisioning in data centers. Recently, Long Short-Term Memory (LSTM) network has been used for host load prediction. However, learning complex noisy variations in host load is still an issue that needs to be addressed. In this work, we propose pCNN-LSTM, a hybrid prediction approach comprising of 1-dimensional Convolution Neural Networks (1D CNN) and LSTM, to predict CPU utilization on Cloud servers at multiple consecutive time-steps. It consists of three parallel dilated 1D CNN layers with different dilation rates for pattern extraction from noisy host CPU usage and an LSTM layer that learns temporal dependencies within the raw usage values as well as within the patterns extracted by the 1D CNN layers. Convolutions with different dilation rates enable the model to learn CPU load variations at different scales. Prediction skill of pCNN-LSTM is demonstrated using Google cluster trace, Alibaba trace and Bitbrains data, and performance is measured using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). pCNN-LSTM achieves up to 15%, 13% and 16% improvements in host load prediction with Google Trace, Alibaba trace and Bitbrains data set, respectively, over LSTM, Bidirectional LSTM (BLSTM), CNN-LSTM, CNN-BLSTM and two of its variants, showing the effectiveness of multi-scale learning capability of pCNN-LSTM and establishes its applicability as an adaptive prediction method for improved capacity planning and provisioning. [ABSTRACT FROM AUTHOR]

Usually, a large number of concurrent bag-of-tasks (BoTs) application execution requests are submitted to cloud data centers (CDCs), which needs to be optimally scheduled on the physical cloud resources to obtain maximal performance. In the current paper, NP-Hard cloud task scheduling (CTS) problem for scheduling concurrent BoT applications is modeled as an optimization problem involving minimization of makespan and energy consumption. Whale optimization algorithm (WOA) has been found effective in solving a wide range of optimization problems. However, standard WOA has certain deficiencies such as inadequate exploration ability, slow convergence, high computation complexity, and insufficient exploration–exploitation phase trade-off. These deficiencies eventually result in unacceptable results when the original WOA is applied over task scheduling optimization problems. To address these limitations, a multi-objective scheduling algorithm called OWPSO is suggested, which incorporates opposition-based learning (OBL) and particle swarm optimization (PSO) mechanisms into the standard WOA method. Firstly, the OBL method is applied to produce an optimal initial population to enhance the exploration and convergence speed of the proposed OWPSO approach in the successive generations. Secondly, PSO and OBL methods are incorporated in the exploration phase of the standard WOA approach to enhance exploration ability further. Thirdly, a fitness-based switching mechanism is added to provide an adequate exploration–exploitation phase trade-off. Finally, a discrete resource allocation heuristic is incorporated in the OWPSO to provide an efficient resource allocation. Simulation experiments over the CloudSim simulator reveal that OWPSO approach results in makespan reduction in the range of 1.68−18.38% (for CEA-Curie workloads), 2.10−24.32% (for HPC2N workloads), and energy consumption reduction in the range of 0.93−14.70% (for CEA-Curie workloads), and 0.73−25.94% (for HPC2N workloads) over other well-known meta-heuristics. Statistical tests and box plots further revealed the robustness of proposed OWPSO algorithm. [ABSTRACT FROM AUTHOR]

Growing demand for reduced local hardware infrastructure is driving the adoption of Cloud Computing. In the Infrastructure-as-a-Service model, service providers offer virtualized computational resources in the form of virtual machine instances. The existence of a large variety of providers and instances makes the decision-making process a difficult task for users, especially as factors such as the datacenter location - where the virtual machine is hosted - have a direct influence on the price of instances. The same instance may present price differences when hosted in different geographically distributed datacenters and, because of that, the datacenter location needs to be taken into account through the decision-making process. Given this problem, we propose the D-AHP, a methodology to aid decision-making based on Pareto Dominance and Analytic Hierarchy Process (AHP). In the D-AHP, the dominance concept is applied to reduce the number of instances to be compared; the instances selection is based on a set of objectives, while AHP ranks the selected ones from a set of criteria and sub-criteria, among them the datacenter location. The results from case studies show that differences may arise in the results, regarding which instance is more suitable for the user, when considering the datacenter location as a criterion to choose an instance. This fact highlights the need to consider this factor during the process of migrating applications to the Cloud. In addition, Pareto Dominance applied early over the set of total instances has proved to be efficient, once it significantly reduces the number of instances to be compared and ordered by the AHP by excluding instances with less computational resources and higher cost in the decision-making process, mainly for larger application workloads. [ABSTRACT FROM AUTHOR]

Dynamic virtual machine (VM) consolidation is a constructive technique to enhance resource usage and is extensively employed to minimize data centers' energy consumption. However, in the current approaches, consolidation techniques are heavily relied on reducing the actively used physical servers (PMs) based on their current resource utilization without considering future resource demands. Also, many of the reported works for cloud workload prediction applied univariate time series-based forecasting models and neglected the dependency of other resource utilization metrics. Thus, resulting in inaccurate predictions, unnecessary migrations, high migration costs, and increased service level agreement violations (SLAVs) may nullify the consolidation benefits. To efficiently address this issue, we propose a multivariate resource usage prediction-based hotspots and coldspots mitigation approach that considers both the current and future usage of resources with O(sk) time complexity, where s and k denote the number of PMs and VMs, respectively. The proposed technique uses a clustering-based stacked bidirectional (Long Short-Term Memory) LSTM deep learning network to predict the future memory and CPU usage of PMs and VMs with high accuracy and O ((Q (Q + W) ∗ Θ) computational complexity, where Q, W, and Θ represent the number of hidden layer cells, outputs, and training epochs, respectively. Through extensive simulations based on Google's cluster workload traces, we demonstrate that our proposed method obtains substantial improvements in terms of prediction performance, energy-efficiency, actively used PMs, VM migrations, and SLA violations over the benchmark approaches. [ABSTRACT FROM AUTHOR]

High-performance computing requires numerous virtual machines (VMs) to meet the demand for services of the end-users. Such computation generates high intra-network traffic among the services running in different VMs. Thus, to achieve a high throughput of the application, VMs are to be deployed in physical machines in a way that reduces the communication cost and delays and thereby meets the high quality of services demanded by the users. On the other hand, the service providers aim to consolidate VMs in a minimum number of active physical machines to reduce their operational costs. Finding an optimal solution to minimize the cost of either the deployment or communication, by themselves are NP-Hard problems. Moreover, an attempt to optimize one disregarding the other may give a solution that is better in terms of cost but inferior in terms of throughput (or vice versa). Hence, we formulate the VM placement problem as a multi-objective optimization problem and propose an Ant Colony Optimization-based VM consolidation algorithm to find a solution in real-time. We have also presented a performance comparison of the proposed algorithm with existing mono-objective and multi-objective algorithms. [ABSTRACT FROM AUTHOR]

In cloud computing, virtual machine placement (VMP) is an important process that identifies the most appropriate physical machine to host the virtual machines (VMs). Nevertheless, determining how to place VMs within the data center to provide high availability and good performance is a difficult challenge for cloud providers. In this paper, with the goal of optimizing the availability and the energy consumption of the cloud data center, an improved genetic algorithm (I-GA) is proposed to solve VMP problem. This new algorithm presents a virtual hierarchy architecture model to combine with the genetic algorithm. The model is able to achieve a near-optimal solution in resolving the availability and energy consumption concerns by innovating the initial population generation step of the I-GA. Finite element analysis is applied as background and CloudSim is used as the experiment simulation. The simulated results demonstrate the significant improvement of the data center's energy efficiency and the successful maintenance of its high availability. The results are also highly competitive compared to the benchmark results of other VMP algorithms. [ABSTRACT FROM AUTHOR]

The growing interest in cloud-based services has made the cost of energy a major concern for cloud service providers. On the other hand, there is an increasing global concern on the carbon footprints of data centers. Shifting from fossil fuels to renewable energy sources is essential to reduce operational costs in data centers while keeping carbon and greenhouse emissions to a minimum. However, considering dynamic electricity pricing, dynamic PUE of data centers, intermittent renewable energy sources and the heterogeneity in workloads and data centers, determining the best data center to host a given VM request is very challenging. Moreover, the fragmentation caused by the placement which increases the energy cost may add to this complexity. This paper presents GreenPacker, a renewable- and fragmentation-aware geographical load balancing method to minimize the energy cost of data centers. This is done by introducing a cost metric that quantifies the suitability of a data center based on its green energy availability, electricity price, PUE, and the fragmentation caused by the placement. Comprehensive analyses using various workload scenarios show that the proposed algorithm can achieve up to 7.2% energy cost savings while maintaining the quality of service at an acceptable level, compared to the state-of-the-art. [ABSTRACT FROM AUTHOR]

Data centers are critical environments that provide support for a wide range of services and applications, and therefore, there is a demand in order to guarantee high availability and reliability required in these environments. This work proposes a strategy based on models, SLA contracts, maintenance policies and optimization techniques for assessing the cost and availability of electrical infrastructures hosted in data centers. The proposed optimization strategy is based on design of experiments (DoE) and uses the availability importance index in order to detect the equipment that most impacts the system's availability and, thus, to be able to propose improvements. In addition, a hybrid modeling approach that considers the advantages of stochastic Petri nets and reliability block diagrams is adopted to assess availability. To illustrate the applicability of the proposed approach, two case studies were carried out where significant results were obtained. In the first study, where the performance of the proposed strategy was compared with the brute force algorithm, it was possible to obtain results close to the optimum ones in a fraction of the time. For example, brute force demanded more than 100 minutes to be evaluated, while the proposed strategy took only 6 seconds. [ABSTRACT FROM AUTHOR]

Performance differentiation and optimization are major dimensions and critical activities in cloud computing systems with shared execution infrastructures. Supporting these features from the perspective of cloud architecture, related concerns and requirements are important challenges, which need more in-depth research. In this regard, this work investigates the dark dimensions of the problem toward realizing an integrated architecture scheme. Therefore, the main goals of the research are to investigate and analyze the operational and non-operational requirements, concerns and principles for performance differentiation and optimization in cloud systems, design a formal architecture supervisor scheme to support performance differentiation and improvement in cloud data centers, propose a conceptual meta-model for fulfilling prerequisites, work stages, design principles, required components and operational elements with essential interconnections and implement different case studies for applying the proposed scheme to executive scenarios with different operating entities and optimization policies. Empirical results present the applicability and usefulness of the proposed scheme in supporting the performance differentiation and improvement. Proposed scheme supports optimization policies and performance improvement in all implemented cases dynamically, and the related metrics have been improved. Finally, major design considerations and recommendations for moving toward other possible optimizations have been proposed. [ABSTRACT FROM AUTHOR]

With the development of big data technologies, green cloud data centers have become a key factor in academia and industry. An energy-efficient cloud data center can save costs for cloud computing users. However, the problem of virtual machine mapping has always been a core problem. In the most existing research, the energy consumption generated by cloud data centers has become an important bottleneck restricting the technology of cloud computing. This paper establishes a consumption model of cloud data center energy and a virtual machine mapping rule. Based on this, a multi-dimensional double descending maximum padding priority (MD3MP2) virtual machine mapping algorithm is proposed. The algorithm can not only solve the one-dimensional virtual machine mapping problem of homogeneous data centers, but also successfully solve the multi-dimensional virtual machine mapping problem of homogeneous data centers. Finally, the algorithm is compared with four other algorithms. The experimental results show that the MD3MP2 algorithm is better than the compared algorithms. [ABSTRACT FROM AUTHOR]

The advent of virtualization technology has created a huge potential application for cloud computing. In virtualization, a large hardware resource is often broken down into smaller virtual units. These small units are then provisioned to different clients. However, these services need to be provided in such a way that resources are properly utilized. To achieve this, many of the scheduling, allocation, and provisioning issues of data centers are formulated as optimization problems. The virtual machine placement problem (VMPP) is a typical provisioning problem of data centers. In VMPP, several virtual machine requests are to be hosted on physical machines such that a minimum number of physical machines are used. This work proposes a cuckoo search (CS) inspired algorithm for solving the VMPP. To improve the algorithm's performance, new cost and perturbation functions are developed. The proposed method was tested on two well-known benchmark datasets. It outperformed the reordered grouping genetic algorithm, best-fit decreasing, first-fit decreasing, and an earlier CS method called multiCSA. [ABSTRACT FROM AUTHOR]

As an important part of the new infrastructure, the cloud data center is developing rapidly, and its energy consumption problem is becoming more and more prominent. Therefore, research on energy-saving technologies for cloud data centers has attracted widespread attention from academia and industry. Some studies have reviewed energy-saving optimization methods and technologies for data centers, but recently, many state-of-the-art optimization methods of energy consumption and energy-saving technologies have sprung out, which are still worth analyzing and discussing. Depending on the in-depth investigation and analysis of related research status, this article firstly focuses on analyzing and discussing the energy-saving technologies of the two components: IT equipment and cooling systems, both of which bring about the largest energy consumption in cloud data centers. As for IT equipment, its energy-saving technologies mainly include the energy saving of servers, storage systems, and network systems. While as for cooling systems, airflow organization in the computer room, thermal-aware scheduling technology, and other new energy-saving technologies are involved. Secondly, on the basis of analyzing the energy-saving technologies of the two major components, a new optimization scheme of energy consumption for the jointing computing system and cooling system is explained. Throughout this work, various energy-saving strategies and technologies have been described and compared. Finally, the future trends and development directions of energy saving for data centers are further promoted, such as integral optimization of energy consumption jointing multiple components, energy saving using artificial intelligence methods, energy saving based on novel hardware equipment, hybrid cooling energy saving, and comprehensive energy conservation with various energy technologies. [ABSTRACT FROM AUTHOR]

The power of rapid scalability and easy maintainability of cloud services is driving many high-performance computing applications from company server racks into cloud data centers. With the evolution of Graphics Processing Units, composing of an extensive array of parallel computing single-instruction-multiple-data processors are being considered as a platform for high-performance computing because of their high throughput. Many cloud providers have begun offering GPU-enabled services for the users where GPUs are essential (for high computational power) to meet the desired Quality-of-service. Virtual machine placement and load balancing the GPUs in the virtualized environments like the cloud is still an evolving area of research and it is of prime importance to achieve higher resource efficiency and also to save energy. The current VM placement techniques do not consider the impact of VM workload type and GPU memory status on the VM placement decisions. This paper discusses the current issues with the First Fit policy of virtual machine placement used in VMWare Horizon and proposes a GPU-aware VM placement technique for GPU-enabled virtualized environments like cloud data centers. The experiments conducted using the synthetic workloads indicate reduction in the energy consumption, reduction in search space of physical hosts, and the makespan of the system. It also presents a summary of the current challenges for GPU resource management in virtualized environments and specific issues in developing cloud applications targeting GPUs under the virtualization layer. [ABSTRACT FROM AUTHOR]

Caused by massive data, large-scale data centers suffer from insufficient storage capacity. This increases the risk of component damage and disk failure. To enhance storage capacity of data centers, scaling has turned out to be one of the most popular choices. RAID-6 is deployed extensively that has superior reliability, availability, and scalability compared to other RAID levels. However, higher reliability also means expensive parity data update cost. To address this issue, this paper proposes an efficient RAID-6 scaling scheme, HS6, for H-Code. The basic idea of HS6 is reducing the total parity data update cost and saving the total scaling time by optimizing data migration and parity data update. The properties of HS6 can be summarized as follows: (1) it utilizes horizontal data migration to eliminate the horizontal parity data update, and (2) it makes full use of original parity data to cut down the total parity data update cost. Numerical results and experimental data analysis indicate that: (1) HS6 decreases the data migration rate by 76.92% ∼ 95.31%, 55.56% ∼ 84.46%, and 14.29% ∼ 16.67% compared to Round-Robin, Semi-RR, and HCS, (2) HS6 saves the total scaling time by 62.05% ∼ 76.75%, 53.24% ∼ 73.94%, and 3.68% ∼ 6.79% against Round-Robin, Semi-RR, and HCS, respectively, under offline and outperforms Round-Robin by 59.99% ∼ 79.02%, Semi-RR by 53.04% ∼ 76.41%, and HCS by 0.56% ∼ 23.78% in total scaling time under four workloads, and (3) HS6 maintains almost the same user average response time with Round-Robin, Semi-RR, and HCS during scaling and after scaling. [ABSTRACT FROM AUTHOR]

To determine whether sensors should transmit collected images back to the data center, an auditing protocol is desired to check these images before transmission. Since images may contain sensitive information of the environment, privacy is an essential requirement for such an auditing protocol. Moreover, since sensors are typically low-power devices and the data center has to handle a variety of sensors, this auditing protocol should be lightweight. Taking both security and efficiency into account, we propose a novel auditing protocol on images in ambient intelligence systems, called AP-AmI (i.e., Auditing Protocol in Ambient Intelligence Systems). In AP-AmI, we use the local binary pattern technique for extracting features from images and design a novel privacy-preserving Euclid distance computation algorithm for determining whether these collected images are useful. Since these two techniques are both lightweight, AP-AmI can achieve high efficiency. At the same time, since feature vectors cannot be extracted by adversaries, AP-AmI can satisfy the privacy requirement. Experimental results show AP-AmI is feasible for real-world applications. [ABSTRACT FROM AUTHOR]

Fog computing as a complementary paradigm to cloud computing is a heuristic shift in service delivery that promises a leap in efficiency and flexibility for cloud-based Internet of Things applications. The performance characteristics of cloud/fog computing attract significant attention from researchers lately. One of the critical challenges in this field is controlling and reducing the massive amount of energy consumption in the cloudlets while still maintaining the Service Level Agreement's performance requirements. Many virtual machine (VM) allocation and consolidation strategies are investigated to address the challenges mentioned earlier. However, many of the solutions save energy at the cost of performance degradation. This paper proposes a novel multi-step VM allocation algorithm called enhanced performance-to-power ratio for workflow applications "E-PRWA" in cloud/fog environment. The proposed heuristic algorithm strives to achieve a trade-off between node performance and power consumption. Operating machine hosts at the highest performance-to-power ratio can save a tremendous amount of energy without degrading system performance. The proposed model consists of four stages: (a) detecting overutilized or underutilized nodes based on the preferred utilization (PU); (b) VM selection for migration from the overutilized nodes to underutilized nodes; (c) switching off selected underutilized nodes; (d) deploying the migration VMs based on the modified best-fit decreasing algorithm with PPR, latency overhead, and computational cost consideration. Extensive simulation results illustrate that compared with three baseline energy-efficient VM allocation and selection algorithms, E-PRWA can achieve an average of up to 65.41% of energy-saving with fewer migration number in fog computing. [ABSTRACT FROM AUTHOR]