Distributed Cloud Environment Research Articles

VNF placement for service chaining in a distributed cloud environment with multiple stakeholders

The adoption of virtualization technologies in networking is promoting a radical innovation in the way network services are managed and delivered. Indeed, some network services may be provisioned to cope with complex and unpredictable traffic demands by dynamically creating a sequence of Virtual Network Functions (VNFs) and steering traffic flows through them. In this context, the optimized deployment of network services, composed of VNFs that may be instantiated in multiple Data Centers (DCs), is one of the most challenging orchestration target. VNF placement is the problem of choosing the set of optimal locations for a chain of VNFs according to the service request and the current characteristics of available computing resources and network links. With respect to the state of the art, our original contribution reflects a multi-stakeholder perspective (subscriber, service providers, infrastructure providers) in a multi-DC environment. We thus consider the problem of placing VNFs to maximize primarily the number of accepted requests from a set of incoming requests and secondarily the satisfaction of subscribers’ preferences. Our model also allows to differentiate service requests in priority levels and guarantees that Quality of Service objectives for accepted service requests are fulfilled, including also a requirement on network service instantiation time. We provide an integer linear programming formulation of this problem that leverages a layered auxiliary graph built for each request in a set. Experimental evaluation is described in detail and an assessment of the proposed placement approach is performed along three main directions: (i) service acceptance ratio in online and offline placement, (ii) preferences’ satisfaction, and (iii) scalability expressed in terms of computational time. The performance of the approach is also compared to a greedy heuristic.

LP-WSC: a linear programming approach for web service composition in geographically distributed cloud environments

In recent years, cloud computing has emerged as the most popular technologies for accessing and delivering enterprise applications as the services to the end users over the Internet. Since different enterprises may offer web services with various capabilities, these web services can be combined with other to provide the complete functionality of a large software application to meet the users’ requests. Therefore, the service composition as an NP-hard optimization problem to combine the distributed and heterogeneous web services is introduced as a challenging issue. In this work, we propose a linear programming approach to web service composition problem which is called ‘LP-WSC’ to select the most efficient service per request in a geographically distributed cloud environment for improving the quality-of-service criteria. Finally, we evaluate the effectiveness of our approach under three scenarios with varying the number of atomic services per set. The experimental results indicate that the proposed approach significantly reduces the cost of selection and composition of the services and also increases the availability of services and the reliability of the servers compared with the other approaches.

A Constraint-Aware Optimization Method for Concurrency Bug Diagnosis Service in a Distributed Cloud Environment

The advent of cloud computation and big data applications has enabled data access concurrency to be prevalent in the distributed cloud environment. In the meantime, security issue becomes a critical problem for researchers to consider. Concurrency bug diagnosis service is to analyze concurrent software and then reason about concurrency bugs in them. However, frequent context switches in concurrent program execution traces will inevitably impact the service performance. To optimize the service performance, this paper presents a static constraint-aware method to simplify concurrent program buggy traces. First, taking the original buggy trace as the operation object, we calculate the maximal sound dependence relations based on the constraint models. Then, we iteratively check the dependent constraints and move forward current event to extend thread execution intervals. Finally, we obtain the simplified trace that is equivalent to the original buggy trace. To evaluate our approach, we conduct a set of experiments on 12 widely used Java projects. Experimental results show that our approach outperforms other state-of-the-art approaches in terms of execution time.

Optimal selection of VMs for resource task scheduling in geographically distributed clouds using fuzzy c‐mean and MOLP

SummaryBecause of widespread distribution of resources in the geographically distributed cloud environment, optimal selection of virtual machines (VMs) is one of the most important challenges for the structure of the network. This is due to the high number of data centers and VMs with different qualities of service parameters. Because of redundancy in the VMs and the high number of service parameters, optimal selection of VMs is an NP‐hard problem. Therefore, a method is required, which can suggest the best VMs on the basis of the user's request and on the service‐level agreements (SLAs). This study focuses on four important factors in SLAs: cost, response time, availability, and reliability.In this paper, we propose a four‐tier structure, Observe, Orient, Decide, and Act, where (1) Observe is responsible for continuous monitoring users' requests and characteristics of data centers and VMs, (2) Orient is responsible for clustering data centers using fuzzy c‐means and based of the four quality of services (SLA's factors) and then the selection of the most suitable data center cluster for the VM selection, (3) Decide is responsible for making decision on the most suitable VMs using multiobjective linear programming, and (4) Act is responsible for the execution of the decision. The proposed structure was implemented, and its effectiveness was evaluated through considering the number of SLA violations.

A moth‐flame optimization algorithm for web service composition in cloud computing: Simulation and verification

SummaryIn recent years, users are becoming increasingly accustomed to using the Internet to gain software resources in the form of web services provided by information technology organizations. Cloud computing is a service delivery paradigm that shares services and resources to access the web services to the end users over the Internet. In the cloud environment, based on the user's needs, various types of services with similar functionalities but different quality‐of‐service (QoS) criteria can be delivered, which often must be combined to meet the users' requests. The optimal selection and composition of these services are realized as an interesting issue. In this paper, we propose a moth‐flame optimization (MFO) algorithm, which is a novel nature‐inspired metaheuristic paradigm for the web service composition (WSC) problem called “MFO‐WSC,” to improve the QoS criteria in the distributed cloud environment. Also, formal modeling is presented for the QoS‐aware MFO‐WSC algorithm with the model checking approach that receives the particular benefits to collaborate the correctness of the proposed algorithm. The correctness of the proposed behavior model is examined using some logical problems such as deadlock‐free, fairness, and reachability conditions in the new symbolic model verifier model checker. The experimental results indicate the effectiveness of the proposed algorithm in comparison with similar related works.

Application Functions Placement Optimization in a Mobile Distributed Cloud Environment

Distributed Computing in 5G Mobile Networks is a potential requirement for certain applications that depends on low latency and information sharing through or with data information sources. Such applications may be observed as a distributed application. We present a tool and method to optimize the deployment of distributed applications, dividing it into Modules, in a 5G Mobile Network environment. To do so we apply an approximation algorithm for the Path Computation and Function Placement Problem described in [1]. We show that under certain circumstances it is beneficial to deploy parts of such applications in a Cloud Computing environment with Distributed Cloud resources at the Mobile Network Edge. We verify our findings with an example, an Augmented Reality application.

Secure data access control with perception reasoning

In spite of all security issues in the cloud system, distributed cloud environment requires an access control model which should be context aware to handle all issues intelligently. It must include role activation process based on the user's context information. In role activation process, the knowledge of reason used for data collection and usage is declared; this can allow the administrator to declare the policies which are context based. Therefore, there is dynamic activation of role permission due to the association of role with context. The complications in the role based access control model reduced by classifying the users into classes or groups having their own access control standards. Access to specific resources and granting/ denying is based on requesting the user identity. Cloud environments consist of different entities, number of resources and user where general access control model fails to cover all the aspects. Here, in the proposed access control with perception reasoning, entities are extended using Extensible Access Control Mark-up Language (XACML) where trust module monitors the random and dynamic behavior of the user with recognizing and restricting the malicious user for illegal data access. By issuing and identity tag to malicious user includes classification of task and data tag with data in the database.

Система виконання сервісів Укрзалізниці як композитних додатків у розподіленій мережі

The main problems of reorganization and operation of modern services of Ukrzaliznytsia are considered in the` article. Special attention is paid to the necessity of gradual and total transition of considered services to the modern technology of deployment and functioning of corporate services in the form of composite applications on workflow technology that operate on the basis of distributed cloud environment. This structure allows us to increase the processing speed and integrity of the existing information, as well as facilitate access to the service for users. For a safe transition to this model, it was initially proposed to deploy existing services as independent composite attachments based on a private cloud. Further, private clouds are proposed to combine into a hybrid cloud with a common pool of resources. This technology will allow us to introduce a flexible distribution system of existing resources between composite applications. It was proposed to enter an independent unit as responsible for the allocation of resources, which will keep records of resources and their allocation. By accepting a set of tasks for a specific composite application, the cloud resource allocation distribution block performs a cascaded decomposition of the set of resources of the hybrid cloud and distributes the resulting subtasks depending on their computational type.

Decision Making Framework for Decentralized Virtual Machine Placement in Cloud Computing

In distributed cloud environment hosts are configured with Local Resource Monitors (LRM). This LRM monitors the underlying hosts’ resource usage, runs independently and balances the underling host’s load by migrating Virtual Machine (VM) instance. For the dynamic environment, each hosts has varying resource requirement, hosts load cannot remain constant. LRM at each host takes decision for VM migration considering static threshold on its own and other hosts current CPU utilization. This result in chances of getting selected same host for VM placement by multiple hosts to reduce resource usage of underlying hosts. The decision making at each server causes the problem of same host identification by multiple hosts during VM placement and consumes extra CPU power and network bandwidth consumption towards each server. This paper addresses the above said issue by proposing decentralized decision making framework for cloud considering hybrid Peer to Peer (P2P) network topology. Proposed solution results avoiding above said issues and balances the load across servers in DC.

Decentralized Predictive Secure VS Placement in Cloud Environment

Hosts in distributed cloud environment configured with Local Resource Monitors (LRM) that runs autonomously, monitors underlying host’s resource usage and balances underlying host’s resource usage by migrating Virtual Machine (VM) to other hosts. LRM takes decision for VM migration at fixed interval considering own current CPU usage and the other hosts CPU usage. The peer hosts unawares about the decision taken by the other hosts LRM about VS migration. As a result of this, there are chances that the same host selection from multiple hosts during the VS placement. This results into destination host to over utilized or the LRM at destination host initiates the VS migration. Several approaches have been proposed for decentralized VS placement that includes two threshold based VS placement, hypercube based VM placement, Ant Colony based VS placement. These approaches does not considers future behavior of the destination hosts after VS placement. This paper discusses the decentralized peer to peer Virtual Server Placement Approach (DPPVP) that considers host’s current as well future CPU utilization for VS placement. The results shows that the proposed system avoids the over utilization of the destination host and host identification by the multiple hosts during VS placement.

A Power Monitoring System Based on a Multi-Component Power Model

As the increasing IT energy consumption emerged as a prominent issue, computer system energy consumption monitoring and optimization has gradually become a significant research forefront. However, most existing energy monitoring methods are limited to hardware-based measurement or coarse-grained energy consumption estimation. They cannot provide fine-grained energy consumption data (i.e., component energy consumption) and high-scalability for distributed cloud environments. In this article, the authors first study widely-used power models of CPUs, memory and hard disks. Then, following an investigation into disk power behaviors in sequential I/O and random I/O, they propose an improved I/O-mode aware disk power model with multiple variables and thresholds. They developed EnergyMeter, a monitoring software utility that can provide accurate power estimate by exploiting a multi-component power model. Experiments based on PCMark prove that the average error of EnergyMeter is merely 5% under a variety of workloads

Cloud-Centric and Logically Isolated Virtual Network Environment Based on Software-Defined Wide Area Network

Recent development of distributed cloud environments requires advanced network infrastructure in order to facilitate network automation, virtualization, high performance data transfer, and secured access of end-to-end resources across regional boundaries. In order to meet these innovative cloud networking requirements, software-defined wide area network (SD-WAN) is primarily demanded to converge distributed cloud resources (e.g., virtual machines (VMs)) in a programmable and intelligent manner over distant networks. Therefore, this paper proposes a logically isolated networking scheme designed to integrate distributed cloud resources to dynamic and on-demand virtual networking over SD-WAN. The performance evaluation and experimental results of the proposed scheme indicate that virtual network convergence time is minimized in two different network models such as: (1) an operating OpenFlow-oriented SD-WAN infrastructure (KREONET-S) which is deployed on the advanced national research network in Korea, and (2) Mininet-based experimental and emulated networks.

A Distributed Locality-Sensitive Hashing-Based Approach for Cloud Service Recommendation From Multi-Source Data

To maximize the economic benefits, a cloud service provider needs to recommend its services to as many users as possible based on the historical user-service quality data. However, when a cloud platform (e.g., Amazon) intends to make a service recommendation decision, considering only its own user-service quality data is insufficient, because a cloud user may invoke services from multiple distributed cloud platforms (e.g., Amazon and IBM). In this situation, it is promising for Amazon to collaborate with other cloud platforms (e.g., IBM) to utilize the integrated data for the service recommendation to improve the recommendation accuracy. However, two challenges are present in the above-mentioned collaboration process, where we attempt to use multi-source data for the service recommendation. First, protecting users’ privacy is challenging when IBM releases its own data to Amazon. Second, the recommendation efficiency and scalability are often low when the user-service quality data of Amazon and IBM update frequently. Considering these challenges, a privacy-preserving and scalable service recommendation approach based on distributed locality-sensitive hashing, i.e., $\textit {SerRec}_{\textit {distri-LSH}}$ , is proposed in this paper to handle the service recommendation in a distributed cloud environment. Extensive experiments on the WS-DREAM data set validate the feasibility of our approach in terms of service recommendation accuracy, scalability, and privacy preservation.

Scientific Workflow Mining in Clouds

Computing clouds have become the platform of choice for the deployment and execution of scientific workflows. Due to the uncertainty and unpredictability of scientific exploration, the execution plan for a scientific workflow may vary from the definition. It is therefore of great significance to be able to discover actual workflows from execution histories (event logs) to reproduce experimental results and to establish provenance. However, most existing process mining techniques focus on discovering control flow-oriented business processes in a centralized environment, and thus, they are mostly inapplicable to the discovery of data flow-oriented, unstructured scientific workflows in distributed cloud environments. In this paper, we present Scientific Workflow Mining as a Service ( ${\sf {SWMaaS}}$ ) to support both intra-cloud and inter-cloud scientific workflow mining. The approach is implemented as a ${\sf {ProM}}$ plug-in and is evaluated on event logs derived from real-world scientific workflows. Through experimental results, we demonstrate the effectiveness and efficiency of our approach.

AN EFFICIENT DEPTH 1 FIXED TREE CONSISTENCY (D1FTC) METHOD FOR DISTRIBUTED DATA TRANSACTIONS IN CLOUD ENVIRONMENT

Cloud provides everything as a service in the IT market. Industries are started to move their services to cloud environment. Managing data transaction is one of the main issues in cloud and it needs efficient architecture to handle the transactional data. This paper discusses existing consistency methods, two-phase commit protocol and proposes a new consistency method to maintain ACID properties in cloud environment. It elaborates how the proposed Depth 1 Fixed Tree Consistency (D1FTC) method works in cloud environment. This work has done to maintain ACID and minimize the transactional issues in distributed cloud environment.

CSA-WSC: cuckoo search algorithm for web service composition in cloud environments

In recent years, service-based applications are deemed to be one of the new solutions to build an enterprise application system. In order to answer the most demanding needs or adaptations to the needs of changed services quickly, service composition is currently used to exploit the multi-service capabilities in the Information Technology organizations. While web services, which have been independently developed, may not always be compatible with each other, the selection of optimal services and composition of these services are seen as a challenging issue. In this paper, we present cuckoo search algorithm for web service composition problem which is called ‘CSA-WSC’ that provides web service composition to improve the quality of service (QoS) in the distributed cloud environment. The experimental results indicate that the CSA-WSC compared to genetic search skyline network (GS-S-Net) and genetic particle swarm optimization algorithm (GAPSO-WSC) reduces the costs by 7% and responding time by 6%, as two major reasons for the reduction of improvement of the quality of service. It also increases provider availability up to 7.25% and the reliability to 5.5%, as the two important QoS criteria for improving the quality of service.

Distributed asteroid discovery system for large astronomical data

Thanks to the advanced technology in astronomical observations, astronomers have collected massive data sets. One of the research tasks is to discover asteroids based on observational data. However, with the fast-growing volume of astronomical data sets, this task becomes enormously relied on computing power. Therefore, the cloud-enable distributed computation and super powerful computation power can offer a good solution to this field. We adopt the Hough transform to link the sequential DETECTIONs of asteroids, and also design a distributed algorithm to process this task on distributed cloud environment for flexible and efficient calculation. Our designs are developed on MapReduce and Spark frameworks. We also utilizes HDFS and HBase in order to reduce disk I/O overhead and increase reliability and scalability. The system architecture is built on the OpenStack which allocates hardware resources flexibly. Our experiment results show significant improvements of calculation time. We believe this will help astronomers to interactively access and analyze data to discover asteroids. The system can also incrementally update DETECTIONs linking with new observed data. It also provides the visual interface to inspect the linked DETECTIONs of asteroids.

Performance analysis of multi-institutional data sharing in the Clouds4Coordination system

Cloud computing is used extensively in Architecture/ Engineering/ Construction projects for storing data and running simulations on building models (e.g. energy efficiency/environmental impact). With the emergence of multi-Clouds it has become possible to link such systems and create a distributed cloud environment. A multi-Cloud environment enables each organisation involved in a collaborative project to maintain its own computational infrastructure/ system (with the associated data), and not have to migrate to a single cloud environment. Such infrastructure becomes efficacious when multiple individuals and organisations work collaboratively, enabling each individual/ organisation to select a computational infrastructure that most closely matches its requirements. We describe the “Clouds-for-Coordination” system, and provide a use case to demonstrate how such a system can be used in practice. A performance analysis is carried out to demonstrate how effective such a multi-Cloud system can be, reporting “aggregated-time-to-complete” metric over a number of different scenarios.

Optimizing virtual machine placement in distributed clouds with M/M/1 servers

As more and more applications migrate into clouds, the placement of virtual machines for these applications has a significant impact on the performance of cloud systems. A number of virtual machine (VM) placement techniques have been proposed over recent years. However, most of the existing works on VM placement ignore the response latency of the requests from tenants. In this paper, we investigate the techniques of VM placement in distributed clouds with stochastic requests from the tenants. We first model the requests for each application from the corresponding tenant as independent Poisson stream. Moreover, based on the analyses of distributed cloud resources, the VMs with their data nodes are modeled as simple M/M/1 queueing systems. Then, we propose the problems of VM placement with two distinct optimization objectives. For each objective, we present the formal definition and prove its NP-hardness. To deal with them, we propose some algorithms and the performances of them are analysed in each section. For applying to the situation of lacking of resource, we propose two extended algorithms. We conduct abundant simulation experiments in distributed cloud environment to evaluate the performance of our proposed algorithms. The simulation results show that the proposed algorithms can significantly improve the performance of their corresponding objectives.

Privacy-Preserving and Scalable Service Recommendation Based on SimHash in a Distributed Cloud Environment

With the increasing volume of web services in the cloud environment, Collaborative Filtering- (CF-) based service recommendation has become one of the most effective techniques to alleviate the heavy burden on the service selection decisions of a target user. However, the service recommendation bases, that is, historical service usage data, are often distributed in different cloud platforms. Two challenges are present in such a cross-cloud service recommendation scenario. First, a cloud platform is often not willing to share its data to other cloud platforms due to privacy concerns, which decreases the feasibility of cross-cloud service recommendation severely. Second, the historical service usage data recorded in each cloud platform may update over time, which reduces the recommendation scalability significantly. In view of these two challenges, a novel privacy-preserving and scalable service recommendation approach based on SimHash, named SerRecSimHash, is proposed in this paper. Finally, through a set of experiments deployed on a real distributed service quality dataset WS-DREAM, we validate the feasibility of our proposal in terms of recommendation accuracy and efficiency while guaranteeing privacy-preservation.