Data-Driven Optimization for Resource Provision in Non-Cooperative Edge Computing Market

Sampling-Based Approximation Schemes for Capacitated Stochastic Inventory Control Models

We provide a refined convergence analysis for the SAA (sample average approximation) method applied to stochastic optimization problems with either single or mixed CVaR (conditional value-at-risk) measures. Under certain regularity conditions, it is shown that any accumulation point of the weak GKKT (generalized Karush-Kuhn-Tucker) points produced by the SAA method is almost surely a weak stationary point of the original CVaR or mixed CVaR optimization problems. In addition, it is shown that, as the sample size increases, the difference of the optimal values between the SAA problems and the original problem tends to zero with probability approaching one exponentially fast.

After years of development, cloud computing has developed into public cloud, private cloud, community cloud, hybrid cloud and other cloud forms. However, the government, financial and other industries have more concerns about security, operation and maintenance, cost and other issues in the process of cloud. In this regard, the industry, academia and other parties have launched the frontier exploration, give different solutions. To meet users’ new needs, we proposed a new model that combines the advantages of public and private clouds. It is called proprietary cloud. Proprietary cloud provides users with more efficient management by incorporating years of experience of public cloud providers. On the one hand, it meets the rigid needs of users’ independent control and exclusive resources, on the other hand, it reduces the maintenance cost and downtime probability of users, and achieves the desired effect of users. It mainly includes self-built model and exclusive model deployment modes. The main difference is whether to accept public cloud management. This paper introduces and analyzes proprietary cloud in detail from the aspects of deployment architecture, mode differences and common deployment patterns. It concludes that proprietary cloud is an indispensable cloud computing deployment pattern in the future.KeywordsCloud computingDigital transformationProprietary cloud

This paper aims to optimise healthcare resource pre-positioning, patient scheduling, and patient transferring under uncertain demands and stochastic resource consumption. We propose a two-stage stochastic programming model that formulates the patient scheduling problem as a Markov decision process. To address complexities and uncertainties, we use Artificial Intelligence (AI) techniques to improve both model formulation and algorithmic performance. To tackle the limited data challenge, we introduce a Wasserstein distance-based ambiguity set and propose a two-stage distributionally robust optimisation (DRO) model, which derives a deterministic equivalent using the Lagrangian dual of non-anticipativity constraints. The solution process is accelerated with a scenario decomposition approach and the K-means clustering method. Both theoretical and numerical results demonstrate the consistency of the two-stage DRO model with the sample average approximation (SAA) method. The potential of AI to improve the model's performance is evident through the significant reduction in computation time achieved with the K-means clustering approach without compromising solution quality. Compared to the SAA method, the DRO model exhibits a considerable reduction in the waiting penalty cost for out-of-sample cases, ranging from 43.06 % to 81.23 % . Numerical results show that the proposed algorithm outperforms SAA methods and several benchmark policies in terms of computational efficiency and solutions.

Cloud computing has evolved to emerge the most topical IT paradigm in recent times. Cloud computing is rapidly transforming the IT landscape. On a pay-as-you-use basis, cloud consumers can access resources, applications and infrastructure provided by cloud providers. Such access could be in form of applications already deployed by cloud providers for use by the cloud users. It could be in form of the capability to develop and deploy user applications using services of a cloud provider. In addition, massive storage infrastructure is available for database and data supplied by the user. The cloud has several unique architectures and many more are still evolving. The primary ones are the SaaS, PaaS and the IaaS that can be deployed on private, public, community and hybrid clouds. This paper examines present developments in the cloud computing architecture and presents guidance for additional research. Papers published in journals, conferences, white papers were analyzed. The objective of this present work is to identify, examine and explain the current trends and development in cloud computing architecture. However, only 13% of the papers examined discussed Others-as-a-Service, while only 26% of the papers reviewed considered issues relating to the major actors involved in cloud computing. This will beneficial to cloud providers, users, and researchers alike

A company planning to implement a public cloud solution should be aware of the risks associated with cloud computing. This is due to the rapidly evolving threats to cloud security that are emerging at an increasing rate in recent years, also to multi-tenant and open nature of the public cloud. Cloud computing is a powerful technology in IT environment and to take advantage of all the benefits of cloud computing, it is necessary to have security and privacy requirements for the implementation and use of the cloud. A risk analysis must be done initially with a cloud provider in order to identify the existing vulnerabilities and to implement the measures to counter those threats. This paper explores the key issues surrounding cloud security, and passes on valuable guidance about not only information security and broadly risk. Also, it will help providers to make informed security decisions about their diverse cloud set-ups and better understand how to reap all the benefits of cloud without compromising their organizations’ security. As well, it discusses the security requirements, and various risks related to the public cloud.

Cloud technologies are becoming pervasive and available for private companies or public institutions in different flavors, mostly public cloud or private clouds. Our focus in this work is on the usage of public cloud technologies by the most popular sites in the Internet. While some studies have described the nascent landscape of public cloud computing 5 years ago, surprising little effort has been put to study the recent evolution of this domain.Using DNS data that enables us to map domains (e.g., netflix.com) and their subdomains (e.g., api.netflix.com) with the cloud providers actually used, we refresh our understanding of this ecosystem. We focus on the dominant four cloud providers, namely Amazon Web Services, Microsoft Azure, Google Cloud Computing and IBM Bluemix. We demonstrate that cloud penetration has clearly increased since 2013, reaching almost 50% of the top 1000 domains, from the Alexa list. Furthermore, a significant fraction of domains use multiple cloud providers simultaneously. Still, domain owenrs remain cautious when it comes to choose which subdomain is actually hosted in the cloud and only 17.8% on average of the subdomains are actually hosted in the cloud. In terms of performance, preliminary results indicate that hosting a subdomain in the cloud pays off as compared to private hosting with a decrease of application level latency of 28%.

A Little Pessimism Goes a Long Way Data-driven optimization is concerned with finding a decision, using data and perhaps a model, that performs well when it is applied on a new unseen data point. Data-driven optimization is challenging because data are limited or the model is wrong or the environment in which the decision is being applied is different from the one in which the training data were collected. Distributionally robust optimization (DRO), a worst case optimization method for finding decisions that are insensitive to model error, can sometimes but not always deliver a decision that has a larger out-of-sample expected reward than the sample average approximation (SAA). “A Data Driven Approach to Beating SAA out of Sample” by Jun-ya Gotoh, Michael Kim, and Andrew Lim shows that if worst case (DRO) solutions fail at this task, then the solution of a best case distributionally optimistic optimization problem will do the job. As good as this sounds, there is a catch: whereas an optimistic decision might beat SAA, the improvement is very modest and comes at the cost of being much more sensitive to model misspecification than both the SAA and the DRO decisions. Moreover, it is easy to make a mistake: it can be difficult to determine with a modestly sized data set whether the best or worst case solution will have the higher expected reward than SAA. In summary, data driven optimization is a trade-off between maximizing the expected reward and controlling the sensitivity of this expectation to model misspecification. When both are considered, a little bit of pessimism goes a long way.

Despite the promising application of cloud computing within government organisations, determining the barriers to the successful implementation of cloud computing remains a major challenge. Often, government organisations face formidable barriers to fully adopting cloud computing functions. These barriers include organisational culture issues, a weak application of service-level agreements (SLAs), compliance issues, interoperability and integration issues, the lack of an organisational strategy, cost calculation methods, cloud service availability, and cloud data security and privacy, as well as insufficient trust in providers. Tackling these barriers while meeting the specifications of government cloud computing involves architectural requirements and mechanisms. In addition, the nature of cloud ecosystem operation and its alignment with appropriate strategies are known to be complex, also considering the growing desire for governance solutions. Although researchers and organisations have recognised cloud computing obstacles prior to the cloud implementation stage, fewer efforts have been made to investigate cloud barriers in-depth during and after cloud implementation projects or when introducing governance solutions. This might partly be attributable to inadequate enterprise requirements for addressing cloud barriers, the complexity of government cloud specifications, and the lack of involvement by cloud stakeholders, such as providers and consumers. Thus, the aim of this dissertation is to develop a governance method consisting of the generation and testing of architectural requirements to overcome dominant cloud barriers and improve cloud computing functionalities in the implementation stage. To achieve this, the roles of related cloud providers and consumers have been investigated. The study was guided by activities of The Open Group Architecture Framework (TOGAF) Architecture Development Method (ADM) to design related architectural requirements that would provide best practices for the development of the governance method. Using a sequential, exploratory mixed method, multiple qualitative case studies on the role of cloud providers were conducted to explore which and how architectural requirements are suitable and applicable for addressing the identified cloud barriers. Afterwards, a quantitative survey study was performed to examine and validate the findings from the qualitative phase and to discover the role of cloud consumers in tackling cloud barriers. The utilisation of dual perspectives, from cloud providers and consumers, to design and measure the suitability and applicability of the architectural requirements is a substantial theoretical contribution in the field of enterprise information systems and cloud governance. Both cloud providers and consumers in the selected research context balanced their roles in the composition of cloud ecosystem projects, reducing complexity and enabling cooperative solutions for handling cloud barriers with different levels of emphasis and prioritisation. The results of this thesis provide significant guidance in eliciting architectural requirements as well as in their successful contribution to addressing the 10 identified cloud barriers in the selected context. Seeking a cloud provider’s perspective through multiple case studies resulted in 24 confirmed, newly emerged architectural requirements and 48 sub-requirements for best practices to support Information Technology (IT) managers and practitioners during the implementation of government cloud projects. From the point of view of the cloud service consumer, statistical evidence showed that the architectural requirements proved to be relevant and applicable and resulted in the validation of only eight important cloud constructs with a scale of 36 architectural requirements. The results at the governance level also produced a clear method that can enhance the realisation of value when business/IT strategies and EA governance practices both take place in cloud projects. The governance method also consists of a set of architectural requirements to support managing and controlling cloud computing implementation projects with related stakeholders and change management initiatives. The architectural requirements have been modelled via the ArchiMate 3.0 modelling language as unique and beneficial contributions for the enhancement of enterprise architecture practices. This thesis thereby introduces a novel cloud governance method with clear mechanisms and practices to serve as crucial resources for related organisations, decision-makers, business practitioners and researchers.

Cloud computing is a promising technology, where the infrastructure, developing platform, software and storage are delivered as a service. With the development of cloud computing, more and more cloud service providers emerge. However, there are no metrics can be referred to compare these providers, so it is difficult for cloud consumers to select the most reliable providers or resources. Thus we present a platform of Services Level Agreement (SLA) in cloud computing. In the platform, we propose a Reputation System to evaluate the reliability of providers or resources to address this challenge, and we also propose a SLA template pool in order to make the SLA negotiation between cloud providers and cloud consumers become more equitable, transparent and convenient.

The Enterprise Level Security (ELS) model focuses on designing secure, distributed web-based systems starting from basic principles. One area of ELS that poses significant design challenges is protection of web server private keys in a public cloud. Web server private keys are of critical importance because they control who can act as the server to represent the enterprise. This includes responding to requests as well as making requests within the enterprise and to its partners. The cloud provider is not part of this trusted network of servers, so the cloud provider should not have access to server private keys. However, current cloud systems are designed to allow cloud providers free access to server private keys. This paper proposes design solutions to securely manage private keys in a public cloud. An examination of commonly used approaches demonstrates the ease with which cloud providers can currently control server private keys. Two designs are proposed to prevent cloud provider access to keys, and their implementation issues are discussed.

To reflect uncertain data in practical problems, stochastic versions of the mathematical program with complementarity constraints (MPCC) have drawn much attention in the recent literature. Our concern is the detailed analysis of convergence properties of a regularization sample average approximation (SAA) method for solving a stochastic mathematical program with complementarity constraints (SMPCC). The analysis of this regularization method is carried out in three steps: First, the almost sure convergence of optimal solutions of the regularized SAA problem to that of the true problem is established by the notion of epiconvergence in variational analysis. Second, under MPCC-MFCQ, which is weaker than MPCC-LICQ, we show that any accumulation point of Karash-Kuhn-Tucker points of the regularized SAA problem is almost surely a kind of stationary point of SMPCC as the sample size tends to infinity. Finally, some numerical results are reported to show the efficiency of the method proposed.

Accessing data from social networks using cloud computing has been adopted by many IT enthusiasts. It allows easy access to data and easy sending of data for communication purposes. For many social network users, cloud computing can simplify processes and save time and money. This article defines cloud computing and shows how one can quickly, safely and easily access data. It also discusses how cloud computing solutions could be beneficial to libraries in these basic areas: technology and acquisition.Today’s more libraries and social networks are joining the cloud. But despite the numerous social networks and data acquisition sites demand for data are still escalating—as they always will. To be able to access data conveniently today’s data competitiveness users of search engines must carefully scrutinize various libraries as well as social media to ensure that they match key data and communication needs and deliver intended results in the most efficient and cost-effective way. To meet these challenges, IT organizations are increasingly moving away from a device-centric view of IT, to a view that is application-, information- and people-centric. Cloud computing aligns with this new world view. Cloud computing technology is enabling IT to do more with the infrastructure that already exists, as well as adding new ways to expand capacity quickly and economically by using external cloud computing resources. This technology is enabling IT managers to treat infrastructure as a common substrate on which they can provision services to users faster in a much more flexible and cost-effective way –without having to re-design or add to the underlying infrastructure. Given the benefits of cloud computing, its broad appeal is not surprising. However, this new approach does raise some concerns. Chief among them is securing data in the cloud. This white paper will briefly review the basic fundamentals of cloud computing and describe how virtualization are enabling enterprises to move toward a flexible, federated computing model with a cohesive mix of private external and internal cloud environments. It then discusses the security implications of cloud computing and provides insights into addressing them, and offers a view of security best practices to adopt when considering cloud services.Cloud computing has become an increasingly popular means of delivering valuable, IT-enabled business services. Adopting cloud technology can be an affordable way to get access to a dynamically scalable, virtualized computing environment. Optimal IT hardware, software, expertise and infrastructure management resources that may not otherwise be available from a cost perspective can be rapidly deployed and easily scaled. Processes, applications and services can be available on demand, regard-less of the user location or device. The cloud provider is responsible for the environment, so organizations can make use of resources for short periods of time without having to maintain the environment when it is not being used. While cloud computing models are attractive because of their flexibility and cost effectiveness, certain challenges must be addressed in order to provide a viable option to traditional data services. First and foremost is the issue of security. The externalized aspect of outsourcing can make it harder to maintain data integrity and privacy, support data and service availability, demonstrate compliance, and secure highly available access to applications and information. In short, cloud computing can present an added level of risk. Organizations must therefore establish trust relationships with their cloud computing providers and understand risk in terms of how these providers implement, deploy, and manage security on their behalf. Whether delivered as part of the service or as specific components added in, your cloud provider should address the fundamentals of security and risk management through a comprehensive approach.

A note on the sample average approximation method for stochastic mathematical programs with complementarity constraints

In this paper, we investigate scenario generation methods to establish lower bounds on the optimal objective value for stochastic scheduling problems that contain random parameters with continuous distributions. In contrast to the Sample Average Approximation (SAA) approach, which yields probabilistic bound values, we use an alternative bounding method that relies on the ideas of discrete bounding and recursive stratified sampling. Theoretical support is provided for deriving exact lower bounds for both expectation and conditional value-at-risk objectives. We illustrate the use of our method on the single machine total weighted tardiness problem. The results of our numerical investigation demonstrate good properties of our bounding method, compared with the SAA method and an earlier discrete bounding method.