Economics of Grids, Clouds, Systems, and Services

Access Control systems are a key resource in computer security to properly manage the access to digital resources. Blockchain technology, instead, is a novel technology to decentralise the control and management of a shared state, representing anything from a data repository to a distributed virtual machine. We propose to integrate traditional Access Control systems with blockchain technology to allow the combined system to inherit the desirable properties blockchain technology provides, mainly transparency and, consequently, auditability. Depending on the application scenario considered, for some systems it may not be desirable to employ a fully decentralised approach. As such, in this paper we outline how our proposal can be adapted to allow for the minimal possible integration of blockchain technology in a traditional Access Control system. In particular, we consider the scenario where Attribute Managers only may be managed on chain through smart contracts. We provide a proof of concept implementation based on Ethereum, and show its performance through experimental results.

Network Functions Virtualization (NFV) decouples the network package performed by network functions from dedicated hardware appliance by running Virtual Network Functions (VNF) on commercial off-the-shelf hardware. Network operators can create customized network services by chaining multiple VNFs, defining a so-called Service Function Chaining (SFC). Because NFV became technically mature recently, the building of such SFCs still needs in-depth knowledge about NFV technology and its descriptors. Furthermore, there is a lack of tools that help to simplify the creation of SFCs. This paper, introduces GENEVIZ, a tool that provides a user-friendly interface for the creation of new SFCs as well as for importing and adjusting acquired SFCs (e.g., from marketplaces of VNFs), in order to create new SFCs based on existing ones. Therefore, this work addresses as well data integrity and provides the functionality to store and validate SFCs through the use of blockchains. Three case studies are presented to provide evidence of the technical feasibility of the solution proposed.

Nowadays information available on the World Wide Web has reached unprecedented growth and it makes difficult for users to find the most relevant for them. In order to alleviate such issue, Recommender Systems (RSs) have been proposed to collect opinions and preferences about a set of items, process such preferences and build a personalized information access.

While the most part of current RSs exploit centralized architecture to provide the service, in this manuscript we propose an alternative approach for building a general purpose RSs that provides to users with more transparent and decentralized rating strategy. Indeed, the proposed framework is built on top of a Distributed Ledger technology platform that runs without any centralized authority and it supports both decentralized ratings and ranking of different items. A preliminary evaluation on the Ethereum test network demonstrates the feasibility of the framework in terms of performance and cost.

In recent years, the parallel computing community has shown increasing interest in leveraging cloud resources for executing parallel applications. Clouds exhibit several fundamental features of economic value, like on-demand resource provisioning and a pay-per-use model. Additionally, several cloud providers offer their resources with significant discounts; however, possessing limited availability. Such volatile resources are an auspicious opportunity to reduce the costs arising from computations, thus achieving higher cost efficiency. In this paper, we propose a cost model for quantifying the monetary costs of executing parallel applications in cloud environments, leveraging volatile resources. Using this cost model, one is able to determine a configuration of a cloud-based parallel system that minimizes the total costs of executing an application.

Vertical elasticity, the ability to add resources on-the-fly to a virtual machine or container, improves the aggregate benefit clients get from a given cloud hardware, namely the social welfare. To maximize the social welfare in vertical elasticity clouds, mechanisms which elicit resource valuation from clients are required. Full Vickrey-Clarke-Groves (VCG) auctions, which allocate resources to optimize the social welfare, are NP-hard and too computationally-complex for the task. However, VCG-like auctions, which have a reduced bidding language compared with VCG, are fast enough. Such is the Simplified Memory Progressive Second Price Auction (SMPSP). A key problem in VCG-like auctions is that they are not completely truthful, requiring participants, who wish to maximize their profits, to estimate their future bills. Bill estimation is particularly difficult since the bill is governed by other participants’ (changing) private bids.

We present methods to estimate future bills in noisy, changing, VCG-like auction environments. The bound estimation method we present leads to an increase of 3% in the overall social welfare.

Cloud architecture spreads services throughout several levels from user-close edge to deep cloud, and while this allocation of resources offers versatility and power, it also presents a challenge: where should each aspect of the service be located? Related to this question is, who should decide? A user-centric approach would invite input from the user, and our model allows users to formulate preferences and submit these to the operator through a voting process wherein they express their preferences for the quality of the services they use. The outcome of the vote is a selection of services and related quality levels which receive preferential treatment.

This process is distinctive in that it operates with only partial information, which may be as much information that can be reasonably obtained. At the same time, it blends well with information that an operator can collect, statically or in realtime, for the user as well as from content and/or application providers.

Energy consumption is a critical operational cost for Cloud providers. However, as commercial providers typically use fixed pricing schemes that are oblivious about the energy costs of running virtual machines, clients are not charged according to their actual energy impact. Some works have proposed energy-aware cost models that are able to capture each client’s real energy usage. However, those models cannot be naturally used for pricing Cloud services, as the energy cost is calculated after the termination of the service, and it depends on decisions taken by the provider, such as the actual placement of the client’s virtual machines. For those reasons, a client cannot estimate in advance how much it will pay. This paper presents a pricing model for virtualized Cloud providers that dynamically derives the energy costs per allocation unit and per work unit for each time period. They account for the energy costs of the provider’s static and dynamic energy consumption by sharing out them according to the virtual resource allocation and the real resource usage of running virtual machines for the corresponding time period. Newly arrived clients during that period can use these costs as a baseline to calculate their expenses in advance as a function of the number of requested allocation and work units. Our results show that providers can get comparable revenue to traditional pricing schemes, while offering to the clients more proportional prices than fixed-price models.

Cloud federations have been seen as a possible solution for the volatility in the number of user requests and for the anti-competitive externalities of the economies of scale in the cloud service sector. In order for a federation to exist in the commercial market, an efficient mechanism for resource and revenue sharing is of paramount importance. In this paper, we design the architecture and specify the business logic for the dynamic operation of such federation platforms. The architecture and federation business logic specification include components, a federation SLA management framework, and revenue sharing mechanisms. It can also offer appropriate incentives to cloud providers for joining a federation. With such dynamism in the platform, cloud providers have the ability to automatically form and dissolve federations, to maintain resource compatibility, and to self-adapt to policies for managing contractual and economic relationships between federation members. This helps in streamlining the overall business process without being dependent on existing business relationships between service providers, between service providers of a federation, and between service providers and customers. This can encourage cloud providers to join in and be benefitted from the federation, thereby contributing to moving cloud computing to the next level.

Network neutrality is being discussed worldwide, with different countries applying different policies, some imposing it, others acting against regulation or even repealing it as recently in the USA. The goal of this paper is to model and analyze the interactions of users, content providers, and Internet service providers (ISPs) located in countries with different rules.

To do so, we build a simple two-regions game-theoretic model and focus on two scenarios of net neutrality relaxation in one region while it remains enforced in the other one. In a first scenario, from an initial situation where both regions offer the same basic quality, one region allows ISPs to offer fast lanes for a premium while still guaranteeing the basic service; in a second scenario the ISPs in both regions play a game on quality, with only one possible quality in the neutral region, and two in the non-neutral one but with a regulated quality ratio between those.

Our numerical experiments lead to very different outcomes, with the first scenario benefiting to all actors (especially the ones in the relaxed-neutrality region) and the second one mainly benefiting mostly to ISPs while Content Providers are worse off, suggesting that regulation should be carefully designed.

The Industrial Internet of Things (IIoT) has become a valuable data source for products and services based on advanced data analytics. However, evidence suggests that industries are suffering a significant loss of value creation from insufficient IIoT data sharing. We argue that the limited utilization of the Sensing as a Service business model is caused by the economic and technological characteristics of sensor data, and the corresponding absence of applicable digital rights management models. Therefore, we propose a property rights enforcement and pricing model by utilizing digital watermarking in combination with product versioning to address the IIoT data sharing incentive problem.

The fast growing worldwide market for Infrastructure as a Service (IaaS) has long been increasingly dominated by the few globally acting hyper-scalers. In turn, the market share and number of small and medium-sized regional IaaS providers have been declining over the past years. This battle for market shares has, however, been astonishingly widely neglected in research. The goal of this paper is therefore to identify and analyze the dominant business model patterns of regional IaaS providers in Germany and compare them regarding their long-term survival prospects. Based on an exploratory multiple-case study with 18 successful regional IaaS providers, two dominant business model patterns were identified: Whereas customizers consciously pursue a busi-ness model being considerably different from the hyperscalers by particularly addressing the discrepancy between the hyperscalers’ standardized offerings and more individual customer requirements, superscalers exhibit several similarities with the hyperscalers and thus act in direct competition. Due to a missing unique selling proposition, except the guaranteed sole data storage in Germany, but at a higher price, superscalers might fall victim to the market consolidation significantly stronger. While scholars obtain a first classification schema of regional IaaS providers which opens up fruitful areas for future research, practitioners get inspirations for their business model innovation process.

Software licensing is changing how organizations and individuals use software. Globally, technical and economic needs affect licensing in many ways and thus creating licensing models and techniques that reflect and serve organizations’ needs, becomes an increasingly challenge in the Network Function Virtualization concept. While NFV continues emerge, it also becomes increasingly important to monitor and manage software licenses. Therefore, in this paper, a license-based architecture is introduced which aims at linking the Network Services with license models throughout SLAs. Specifically, we have introduced an interconnection between license models and SLAs, in which we aim at an efficient and flexible service orchestration in a beyond MANO SP.

Current projects applying blockchain technology to enhance the trust of NFV environments do not consider the VNF repository. However, the blockchain’s properties can enhance trust by allowing to verify a VNF package’s integrity without relying (a) on a Trusted Third Party (TTP) for remote attestation or (b) a secure database. This paper presents BUNKER, a Blockchain-based trUsted VNF packagE Repository, intended to be integrated with traditional database-based package verification environments, acting as a trusted repository containing VNF package information. Moreover, BUNKER allows users to acquire VNFs without the need of a TTP using an Ethereum Smart Contract (SC). The SC automatically transfers license fees to the vendor once a VNF is acquired, and sends the VNF package’s link to the buyer before verifying its integrity.

With the growing concerns regarding energy consumption, companies and industries worldwide are looking for ways to reduce their costs and carbon footprint linked to energy usage. The rising cost of energy makes energy saving and optimisation a real stake for businesses which have started to implement more intelligent energy management techniques to achieve a reduction of costs. As industries migrate towards more renewable energy sources and more sustainable consumption models, decentralised energy infrastructure is required where actors can manage and monetise energy capabilities.

In fish processing industries, energy is utilised to operate a range of cold rooms and freer units to store and process fish. Modelling thermal loads, appliance scheduling and integration of renewable energy represent key aspects in such industries. To enable the transition towards Industry 4.0 and to efficiently optimise energy in fish industries, multi-agent systems can provide the mechanisms for managing energy consumption and production with standalone entities that can interact and exchange energy with a view of achieving more flexible and informed energy use.

In this paper, we propose a multi-agent coordination framework for managing energy in the fish processing industry. We demonstrate how agents can be devised to model appliances and buildings and to support the formation of smart energy clusters. We validate our research based on a real use-case scenario in Milford Haven port in South Wales by showing how multi-agent systems can be implemented and tested for a real fish industrial site.

Cyber-Physical Systems (CPSs) comprise multiple cyberparts, physical processes, and human participants (end-users) that affect them, and vice versa. During the design of such systems, it is critical for the designer to take into account the end-user-perceived quality of provided services, as well as their cost, and integrate them into the CPSs; striking a satisfactory balance between quality and affordability is critical to system acceptance. In this work, we propose a model-based approach, using the Systems Modeling Language (SysML), to explore system design, encapsulating Quality of Service (QoS) and cost aspects, as system requirements, into a core model. Via this approach, the designer can define the system structure, configure it, measure and evaluate the quality, while analyzing cost, and find the best solution(s) for a correct design. As a use case, this approach is applied to a healthcare CPS, namely the Remote Elderly Monitoring System (REMS). In that context, managing REMS QoS and cost requirements, can contribute to an effective system design and implementation, enhancing the end-user satisfaction.

Enterprises have been challenged to adopt practices of sustainability to benefit shareholders and society with goods standing much beyond monetary profit or required by law. In combination with environmental and economic concerns, Corporate Social Responsibility (CSR) has become an option to leverage businesses with good reputation and to attract sustainability-aware market segments. In line with such a demand, this paper presents a systematic literature review of conceptual modeling studies referring explicitly to certifications, laws or norms of CSR. The more specific research goal of this work is to discover ontologies for representing CSR best practices, design patterns or policies. In total, 921 peer-reviewed papers were analyzed, from which only 17 were considered relevant for data extraction. The main result of this work is the identification of a research gap in explicit knowledge representation of CSR practices for Information Systems design, which ought to be filled to complement the (dominant) economic perspective on sustainability.

We consider the optimization problem of a multi-resource, multi-unit VCG auction that produces an exact, i.e., non-approximated, social welfare. We present an algorithm that solves this optimization problem with pseudo-polynomial complexity and demonstrate its efficiency via our implementation. Our implementation is efficient enough to be deployed in real systems to allocate computing resources in fine time-granularity. Our algorithm has a pseudo-near-linear time complexity on average (over all possible realistic inputs) with respect to the number of clients and the number of possible unit allocations. In the worst case, it is quadratic with respect to the number of possible allocations. Our experiments validate our analysis and show near-linear complexity. This is in contrast to the unbounded, nonpolynomial complexity of known solutions, which do not scale well for a large number of agents.

For a single resource and concave valuations, our algorithm reproduces the results of a well-known algorithm. It does so, however, without subjecting the valuations to any restrictions and supports a multiple resource auction, which improves the social welfare over a combination of single-resource auctions by a factor of 2.5-50. This makes our algorithm applicable to real clients in a real system.

In this paper, we present Optimisation as a Service (OaaS) for an integrated process planning and scheduling in smart factories based on a distributed multi-criteria genetic algorithm (GA). In contrast to the traditional distributed GA following the island model, the proposed islands are executed asynchronously and exchange solutions at time points depending solely on the optimisation progress at each island. Several solutions’ exchange strategies are proposed, implemented in Amazon Elastic Container Service for Kubernetes (Amazon EKS) and evaluated using a real-world manufacturing problem.

In this paper, we presented a comprehensive stability analysis of statistical models derived from the network usage data to design an efficient and optimal resource management in a Cloud data centre. In recent years, it has been noticed that network has a significant impact on the HPC and business critical applications when they are run in a cloud environment. The existing VM placement algorithms lack capabilities to deploy such applications in an effective way and cause performance degradation. As a result, there is an urge for a network-aware VM placement algorithm which will consider the application behaviour and system capability. Our approach uses static models based on simple probability distribution concept and partition (number theory) to characterise and predict the resource usage behaviour of the VMs. However, the stability of those models is a key requirement to ensure a persistent placement of the VMs which can prevent their frequent migration and keep the infrastructure rigid. The paper investigates the stability of the models with respect to time. Sticky HDP-HMM method was proven highly capable to model the monitoring data with a certain accuracy. The refined data was further used to estimate the resource consumption of each VM and physical host running in the infrastructure. A stability parameter has been defined to determine the level of steadiness of the models that gives us a clear indication on whether the models can be used further to derive an optimal placement decision for new VMs. The paper ends with a discussion on instance based stability analysis and future work.

The regional licensing of 5G spectrum represents an exemplary use case for the huge potential of a geographically fine-grained spectrum assignment. It raises the question, whether new types of spectrum licenses could augment the current arsenal of static allocation schemes with central control. In our contribution, we introduce the use of genetic algorithms (GA) as integral part of ex-ante evaluations of new licensing schemes. We apply the method to calculate the near-optimal exploitation of available spectrum resources in the course of an academic simulation. It assesses the effects of interference-based sub-licensing contracts, also known as Pluralistic Licensing Contracts (PLCs), on mobile networks. Our findings suggest that PLCs might provide a low-effort exploitation of underutilized spectrum reserves, e.g. in sparse user populations, and constitute a highly scalable means for the pricing of externalities.

Cloud TV will play an important role in future pay-TV services and is quickly becoming the next arena for TV content providers. This emphasizes the need for a technology roadmap to address several key issues that may affect the deployment of future Cloud TV services. Taking into account an important blend of social, economic and technological factors, three alternative technologies, Internet Protocol TV, Over the Top and Smart TV have been investigated and ranked using the Analytical Hierarchy Process. The results reveal that OTT seems to take the precedence and security, privacy, accessibility, costs saving and time-to-market are crucial aspects, need to be taken into account.

The digital world demands a network infrastructure to supply connectivity to any participant anywhere. Sustainable networks require balanced value flows. Value is connectivity delivered at a material and service cost to compensate, involving diverse participants, ranging from consumers to providers, such as last mile access, regional transport, Internet carriers, or content providers. We focus on the case of wireless mesh networks that deliver connectivity through access points and a mesh network that routes traffic to Internet gateways, provisioned by several device owners and service operators [1‐3].

The presented work is motivated by the need for balance and automation among services delivered, costs and incentives for participation in these decentralised networks. This balance is key for achieving extensible network infrastructures that can deliver widespread availability of Internet connectivity with minimal barriers of entry.

The simplification of resource management for container is one of the most important services of Kubernetes. However, the simplification of distributed provisioning and scheduling decisions can impact significantly in cost outcomes. From an economic point of view, the most important factor to consider in container management is performance interference among containers executing in the same node. We propose a model driven approach to improve resource usage in overall deployment of applications. Petri Net models, a Confirmatory Factor Analysis (CFA)-based model and a regression model allows to predict performance degradation of the execution of containers in applications. Time series indices can provide an accurate enough characterisation of the performance variations in the execution lifetime of applications. These indices can be used in new scheduling strategies to reduce the number of resources used in shared cloud environments as Kubernetes.