Multi-cloud Scenario Research Articles

Dynamic pricing on maximum concurrency for heterogeneous instances using hyperparameter optimization in dueling deep reinforcement learning in a multi-cloud scenario

Users possess the option to rent instances of various sorts, in a variety of regions, and a variety of availability zones, thanks to cloud service carriers like AWS, GCP, and Azure. In the cloud business right now, fixed price models are king when it comes to pricing. However, as the diversity of cloud providers and users grows, this approach is unable to accurately reflect the market’s current needs for cost savings. As a consequence, a dynamic pricing strategy has become a desirable tactic to better handle the erratic cloud demand. In this study, a deep learning model was used to propose a dynamic pricing structure that ensures service providers are treated fairly in a multi-cloud context. The computational optimization of DL approaches can be severely hampered by the requirement for human hyperparameter selection. Traditional automated solutions to this issue have inadequate durability or fail in specific circumstances. To choose the hyper-parameters in the Dueling Deep Q-Network (DDQN), the hybrid DL approach in this study uses the concept-based wild horse optimization (WHO) method. A community of untamed horses is evolved, and the fitness of the population is evaluated concurrently to estimate the optimum hyper-parameters. The plan changes the price appropriately to promote the use of underutilized resources and discourage the use of overutilized resources. The evaluation’s findings demonstrated that the suggested strategy can lower end-user costs while conducting compute- and data-intensive activities in a multi-cloud environment. The research was concluded by comparing current models after the results were analyzed using various performance indicators.

The Application of Optimization Algorithms for Workflow Scheduling Based on Cloud Computing IaaS Environment in Industry Multi-Cloud Scenarios

The Application of Optimization Algorithms for Workflow Scheduling Based on Cloud Computing IaaS Environment in Industry Multi-Cloud Scenarios

Reliability-aware web service composition with cost minimization perspective: a multi-objective particle swarm optimization model in multi-cloud scenarios

Reliability-aware web service composition with cost minimization perspective: a multi-objective particle swarm optimization model in multi-cloud scenarios

Dim and Small Target Detection Based on Energy Sensing of Local Multi-Directional Gradient Information

It is difficult for traditional algorithms to remove cloud edge contours in multi-cloud scenarios. In order to improve the detection ability of dim and small targets in complex edge contour scenes, this paper proposes a new dim and small target detection algorithm based on local multi-directional gradient information energy perception. Herein, based on the information difference between the target area and the background area in the four direction neighborhood blocks, an energy enhancement model for multi-directional gray aggregation (EMDGA) is constructed to preliminarily enhance the target signal. Subsequently, a local multi-directional gradient reciprocal background suppression model (LMDGR) was constructed to model the background of the image. Furthermore, this paper proposes a multi-directional gradient scale segmentation model (MDGSS) to obtain candidate target points and then combines the proposed multi-frame energy-sensing (MFESD) detection algorithm to extract the true targets from sequence images. Finally, in order to better illustrate the effect of the algorithm proposed in this paper in detecting small targets in a cloudy background, four sequence images are selected for detection. The experimental results show that the proposed algorithm can effectively suppress the edge contour of complex clouds compared with the traditional algorithm. When the false alarm rate Pf is 0.005%, the detection rate Pd is greater than 95%.

Towards securing cloud data in the multi-cloud scenario

Cloud computing has emerged to be the accepted computing model which provides services on-demand. The most used service layer is infrastructure as a Service (IaaS) to outsource data to the cloud. With this service, organizations and individuals can avail of cloud services in pay as you use fashion instead of investing money for such infrastructure. Cloud provides many such benefits to its users. However, as the cloud servers are remote and assumed to be untrusted, users are worried about data security. Initially, a single cloud was used to store data. With the advancements in technologies and for reliability reasons, the concept of multi-cloud has emerged. The security and reliability issues with a single cloud can be overcome with multi-cloud systems. The rationale behind this is that a single cloud might have malicious insiders. When two or more clouds collaborate and provide services to end-users, it is expected to have more reliability and possible reduction in malicious insiders. This paper focuses on studying the potential security of data that is stored in multi-cloud. We built an algorithm and prototype application that demonstrates the concept of securing data in a multi-cloud environment. The empirical results revealed that the proposed system could ensure the data outsourced to cloud computing where a multi-cloud scenario prevails.

Towards securing cloud data in the multi-cloud scenario

Cloud computing has emerged to be the accepted computing model which provides services on-demand. The most used service layer is infrastructure as a Service (IaaS) to outsource data to the cloud. With this service, organizations and individuals can avail of cloud services in pay as you use fashion instead of investing money for such infrastructure. Cloud provides many such benefits to its users. However, as the cloud servers are remote and assumed to be untrusted, users are worried about data security. Initially, a single cloud was used to store data. With the advancements in technologies and for reliability reasons, the concept of multi-cloud has emerged. The security and reliability issues with a single cloud can be overcome with multi-cloud systems. The rationale behind this is that a single cloud might have malicious insiders. When two or more clouds collaborate and provide services to end-users, it is expected to have more reliability and possible reduction in malicious insiders. This paper focuses on studying the potential security of data that is stored in multi-cloud. We built an algorithm and prototype application that demonstrates the concept of securing data in a multi-cloud environment. The empirical results revealed that the proposed system could ensure the data outsourced to cloud computing where a multi-cloud scenario prevails.

Security and Privacy in Multi-Cloud Environments

In the modern-day panorama of computing, the adoption of multi-cloud environments has emerge as increasingly more commonplace, providing businesses stronger flexibility, scalability, and resource optimization. However, this proliferation introduces complex demanding situations related to safety and privateness. This research makes a speciality of examining the intricacies of protection and privacy concerns within multi-cloud setups, mainly addressing demanding situations and featuring solutions inside the geographical regions of facts encryption, identification control, and compliance. A important mission in multi-cloud environments is the implementation of powerful statistics encryption strategies throughout numerous systems. Protecting data confidentiality and integrity as it traverses distinctive cloud offerings is of paramount importance. This have a look at seriously evaluates present encryption methodologies, assesses their suitability in multi-cloud scenarios, and indicates progressive tactics to reinforce records safety. Identity control emerges as another important factor of the multi-cloud safety paradigm. The secure orchestration of user identities throughout disparate cloud services needs cautious exam. This studies explores the complexities surrounding identification federation, authentication mechanisms, and get entry to controls inside multi-cloud environments. By figuring out vulnerabilities and proposing strong identity management solutions, the study contributes to the improvement of secure multi-cloud infrastructures. Additionally, making sure compliance with regulatory frameworks and industry standards poses a powerful venture in multi-cloud protection. The examine scrutinizes the complex panorama of compliance necessities, encompassing diverse geographical and enterprise-specific policies. By analyzing the gaps and intricacies in present compliance frameworks, the research proposes a comprehensive method to make sure adherence to guidelines whilst operating in a multi-cloud environment. In conclusion, this studies addresses the challenges and solutions inherent in ensuring strong protection and privacy measures within multi-cloud environments. By specializing in facts encryption, identity management, and compliance, the observe provides valuable insights for companies navigating the complex terrain of multi-cloud computing.

Novel protected sub-frame selection based interference mitigation and resource assignment in heterogeneous multi-cloud radio access networks

The future network in a real environment is assumed to involve a radio access network comprising several clouds, as opposed to the single-cloud scenario in the recent cloud radio access network (CRAN) literature. Hence, this research presents the more favorable multi-cloud scenario to address the limited coverage and processing abilities of CRAN, in which inter-tier and inter-cloud interferences are considered and resource allocation is presented to enhance both the spectral and energy efficiencies. In order to mitigate interferences and enhance the spectral and energy efficiency (EE), we adopted protective sub-frame (PSF) based interference mitigation. To deal with that, part of the total sub-frames construct PSF matrix from which the combinations of sub-channels are selected to communicate with the edge low quality of service (QoS) users. In line with this, we introduced a novel sub-frame selection algorithm using the concept of cooperative game of learning automata, in which UEs can learn past sub-frame selection history from other neighboring UEs and updated their probability as indicative of the stochastic characteristics of the selected sub-frame to influence the next sub-frame selection. Accordingly, power allocation for the selected combination of sub-frames is formulated to allocate transmission power based on the received SINR. Furthermore, the objective problem is formulated as a non-convex energy-efficient resource assignment problem and the non-convex problem is efficiently converted to convex feasible problem utilizing the nonlinear fractional programming, upon which we then develop an efficient iterative algorithm. Finally, we evaluated the performance of our proposed model in terms of convergence, EE, and overall system capacity. For that, simulation is conducted and results confirm that the corresponding cooperative learning automata-based sub-frame selection enhances the EE significantly with faster convergence and improved system capacity.

Orchestrating the Deployment of High Availability Services on Multi-zone and Multi-cloud Scenarios

Cloud computing has become one of the most used platforms to deploy High Availability (HA) solutions for its flexibility, on-demand provisioning, and elasticity. However, although many providers offer specific tools for HA support, like floating IPs and load balancing, the analysis of downtime at public cloud providers in previous years shows that a combination of several availability zones or cloud providers is required to achieve “five nines” availability. Besides reducing the chances of failure, the use of multiple availability zones and geographically distributed clouds may additionally bring performance and cost benefits. However, the orchestration, in an efficient and adaptive way, of HA multi-tier services in multi-zone and multi-cloud environments brings several challenges. This paper presents a novel orchestration method to automate the deployment and management of high availability multi-tier services on multiple availability zones, by introducing new affinity mechanisms, such as VM to location and role to role affinity/anti-affinity rules. Furthermore, we also extend this solution to multi-cloud scenarios, based on the replication or distribution of the service components among various clouds, along with their corresponding affinity rules.

Optimal virtual network function placement in multi-cloud service function chaining architecture

Service Function Chaining (SFC) is the problem of deploying various network service instances over geographically distributed data centers and providing inter-connectivity among them. The goal is to enable the network traffic to flow smoothly through the underlying network, resulting in an optimal quality of experience to the end-users. Proper chaining of network functions leads to optimal utilization of distributed resources. This has been a de-facto model in the telecom industry with network functions deployed over underlying hardware. Though this model has served the telecom industry well so far, it has been adapted mostly to suit the static behavior of network services and service demands due to the deployment of the services directly over physical resources. This results in network ossification with larger delays to the end-users, especially with the data-centric model in which the computational resources are moving closer to end users. A novel networking paradigm, Network Function Virtualization (NFV), meets the user demands dynamically and reduces operational expenses (OpEx) and capital expenditures (CapEx), by implementing network functions in the software layer known as virtual network functions (VNFs). VNFs are then interconnected to form a complete end-to-end service, also known as service function chains (SFCs). In this work, we study the problem of deploying service function chains over network function virtualized architecture. Specifically, we study virtual network function placement problem for the optimal SFC formation across geographically distributed clouds. We set up the problem of minimizing inter-cloud traffic and response time in a multi-cloud scenario as an ILP optimization problem, along with important constraints such as total deployment costs and service level agreements (SLAs). We consider link delays and computational delays in our model. The link queues are modeled as M/D/1 (single server/Poisson arrival/deterministic service times) and server queues as M/M/1 (single server/Poisson arrival/exponential service times) based on the statistical analysis. In addition, we present a novel affinity-based approach (ABA) to solve the problem for larger networks. We provide a performance comparison between the proposed heuristic and simple greedy approach (SGA) used in the state-of-the-art systems. Greedy approach has already been widely studied in the literature for the VM placement problem. Especially we compare our proposed heuristic with a greedy approach using first-fit decreasing (FFD) method. By observing the results, we conclude that the affinity-based approach for placing the service functions in the network produces better results compared against the simple greedy (FFD) approach in terms of both, total delays and total resource cost. We observe that with a little compromise (gap of less than 10% of the optimal) in the solution quality (total delays and cost), affinity-based heuristic can solve the larger problem more quickly than ILP.

Combining reputation and QoS measures to improve cloud service composition

In order to provide a wide range of composite Cloud services, providers need to establish mutual agreements on large-scale distributed multi-cloud scenario. In such a way, providers can compose effective and efficient service workflows by taking resources of their own competitors and gain the capability to satisfy unexpected workload peaks. In this paper, we propose a reputation-based model capable to support the composition of complex Cloud services by taking into account both costs and measures of QoS which are collected by measuring both system measures and reputation feedbacks provided by the customers. The proposed model has been validated by a set of experimental results obtained by means of a number of simulations.

Identity-based provable data possession revisited: Security analysis and generic construction

Provable Data Possession (PDP), which enables cloud users to verify the data integrity without retrieving the entire file, is highly essential for cloud storage. Observing all the existing PDP schemes rely on the Public Key Infrastructure (PKI), Wang proposed an identity-based distributed provable data possession (ID-DPDP) scheme that can (1) eliminate the complex certificate management and (2) be applied to the multi-cloud scenario. The scheme is efficient, flexible and supports private verification, delegated verification and public verification. In this paper, we find that ID-DPDP is flawed since it fails to achieve soundness. We then fix the flaw by presenting a generic construction for identity-based PDP (ID-PDP) protocol, derived from secure digital signature schemes and traditional PDP protocols. We prove that the soundness of the generic ID-PDP construction depends on the security of the underlying PDP protocols and the signature schemes. An instance of the generic construction by utilizing a state-of-the-art PDP protocol due to Shacham and Waters and BLS short signature scheme is given. Moreover, a new ID-DPDP protocol is obtained by extending the basic ID-PDP to multiple clouds environment. The implementation shows that the proposed ID-PDP protocol is efficient.

An approximate dynamic programming approach to resource management in multi-cloud scenarios

ABSTRACTThe programmability and the virtualisation of network resources are crucial to deploy scalable Information and Communications Technology (ICT) services. The increasing demand of cloud services, mainly devoted to the storage and computing, requires a new functional element, the Cloud Management Broker (CMB), aimed at managing multiple cloud resources to meet the customers’ requirements and, simultaneously, to optimise their usage. This paper proposes a multi-cloud resource allocation algorithm that manages the resource requests with the aim of maximising the CMB revenue over time. The algorithm is based on Markov decision process modelling and relies on reinforcement learning techniques to find online an approximate solution.

Resource Allocation on a Hybrid Cloud for Smart Grids

The use of hybrid clouds enables companies to cover their demands of IT resources saving costs and gaining flexibility in the deployment of infrastructures by paying under demand these resources. However, considering a scenario with various services to be allocated in more than one cloud, it is necessary to find the distribution of services that minimizes the overall operating costs. This paper researches on the resource allocation methodology to be applied in a multi-cloud scenario based on the findings derived from the framework used for the FINESCE project. The purpose of this work is to define a methodology to assist on the hybrid cloud selection and configuration in the Smart Grid for both generic and highly-constrained scenarios in terms of latency and availability. Specifically, the presented method is aimed to determine which is the best cloud to allocate a resource by (1) optimizing the system with the information of the network and (2) minimizing the occurrence of collapsed or underused virtual machines. Also, to assess the performance of this method and any alternative proposals, a general set of metrics has been defined. These metrics have been refined taking into account the expertise of FINESCE partners in order to shape Smart Grid clouds and reduce the complexity of computation. Finally, using the data extracted from the FINESCE testbed, a decision tree is used to come up with the best resource allocation scheme.

The Evolution of Cloud Computing in ATLAS

The ATLAS experiment at the LHC has successfully incorporated cloud computing technology and cloud resources into its primarily grid-based model of distributed computing. Cloud R&D activities continue to mature and transition into stable production systems, while ongoing evolutionary changes are still needed to adapt and refine the approaches used, in response to changes in prevailing cloud technology. In addition, completely new developments are needed to handle emerging requirements. This paper describes the overall evolution of cloud computing in ATLAS. The current status of the virtual machine (VM) management systems used for harnessing Infrastructure as a Service resources are discussed. Monitoring and accounting systems tailored for clouds are needed to complete the integration of cloud resources within ATLAS' distributed computing framework. We are developing and deploying new solutions to address the challenge of operation in a geographically distributed multi-cloud scenario, including a system for managing VM images across multiple clouds, a system for dynamic location-based discovery of caching proxy servers, and the usage of a data federation to unify the worldwide grid of storage elements into a single namespace and access point. The usage of the experiment's high level trigger farm for Monte Carlo production, in a specialized cloud environment, is presented. Finally, we evaluate and compare the performance of commercial clouds using several benchmarks.

Recent developments in high‐performance computing and simulation: distributed systems, architectures, algorithms, and applications

Recent developments in high‐performance computing and simulation: distributed systems, architectures, algorithms, and applications

A user-centric multi-PaaS application management solution for hybrid multi-Cloud scenarios

Cloud Platform as a Service (PaaS) is a rapidly growing IT paradigm which enables software developers to deploy applications without the burden of software platform maintenance. Currently, the PaaS market is dominated by a few providers that promote incompatible standards. This introduces adoption barriers that prevent the interoperability between heterogeneous PaaS offerings, so software developers are not able to manage distributed applications spanning multiple public/private clouds. In this paper we present a multi-PaaS application management solution as a result of the Cloud4SOA European project that addresses these challenges. To clarify this approach a distributed deployment and cloud bursting scenarios are used.