CLOSER 2019 Abstracts

Area 1 - Cloud Computing Fundamentals
Full Papers Short Papers
Area 2 - Cloud Operations
Short Papers
Area 3 - Data as a Service
Short Papers
Area 4 - Edge Cloud and Fog Computing
Full Papers Short Papers
Area 5 - Mobile Cloud Computing
Short Papers
Area 6 - Service Modelling and Analytics
Full Papers Short Papers
Area 7 - Services Science
Full Papers Short Papers
Area 8 - Cloud Computing Platforms and Applications
Full Papers Short Papers
Area 9 - Cloud Computing Enabling Technology
Full Papers Short Papers

Area 1 - Cloud Computing Fundamentals

Full Papers
Paper Nr: 33
Title:

Continuous and Client-centric Trust Monitoring in Multi-cloud Storage
Authors:

Dimitri Van Landuyt, Luuk Raaijmakers, Ansar Rafique and Wouter Joosen
Abstract: Multi-cloud storage is the practice of composing the data tier of an application with heterogeneous cloud storage technologies, resources and services. In a federated cloud storage architecture which involves multiple cloud storage providers, both the complexity and the importance of trust management increases drastically. A trust relation is established between a data owner and a cloud storage provider when the data owner subscribes to the service and service level agreements (SLAs) are established. In practice, this trust relation is seldom revised, only when serious infractions are discovered and made public. In this paper, we evaluate the potential of continuous and client-centric trust monitoring of cloud storage services. This approach leverages upon the statistical correlations between black-box performance metrics and reported white-box metrics, and identifies significant deviations between both. We evaluate in terms of (a) the effectiveness of correlating black-box and white-box measurements, and (b) the incurred performance overhead of the approach to continuously monitor for trust.
Download
Paper Nr: 69
Title:

Towards Short Test Sequences for Performance Assessment of Elastic Cloud-based Systems
Authors:

Michel Albonico and Paulo Varela
Abstract: Elasticity is one of the main features of cloud-based systems (CBS), helping them to meet performance requirements under a varying workload. Given the great number of combinations among workload and elastic adaptation parameters, assessing their effect on CBS performance may be prohibitive. Existing systematic combinatorial testing approaches can help to reduce such combinations, though most of them only consider conventional software architectures. In the literature, we only find a single work on elastic CBS combinatorial testing, presented by some of the authors. However, the paper only presents experimental results on 2-wise elasticity parameter interactions and shallowly explores the performance issue causes. In this paper, we lead a further experiment by using our previous approach to generate performance test cases that cover three elasticity parameter interactions (i. e., 3-wise), one interaction longer than on the previous paper. Despite the significant increase in execution time and cost, new experimental results do not reveal any new critical performance issue by 3-wise, which enforces the acceptance of 2-wise elasticity parameter interactions.
Download
Paper Nr: 77
Title:

Assuring Cloud QoS through Loop Feedback Controller Assisted Vertical Provisioning
Authors:

Armstrong Goes, Fábio Morais, Eduardo L. Falcão and Andrey Brito
Abstract: A problem observed in cloud environments is the assurance of Quality of Service (QoS) in the execution of applications. In the context of batch applications, a common goal is ensuring deadline compliance. However, estimating the required amount of resources to assure execution deadline may be difficult and prone to underallocations or superallocations. An alternative is modifying the amount of allocated resources in case the progress is not satisfactory, using horizontal or vertical scaling. Following this strategy, this work proposes a provisioning method based on PID controllers and vertical scaling for batch applications. To evaluate the proposed provisioning method on assuring QoS, we executed two microbenchmarks and one Big Data application on a real cloud infrastructure where the provisioning was controlled by our method. Results show that the provisioning method using the Proportional-Derivative controller is effective on ensuring QoS regarding deadline compliance, allocation smoothness and resource efficiency, though it requires additional adjusts when provisioning resources to applications with non-linear progress.
Download
Short Papers
Paper Nr: 6
Title:

Cloud Service Quality Model: A Cloud Service Quality Model based on Customer and Provider Perceptions for Cloud Service Mediation
Authors:

Claudio Giovanoli
Abstract: The field of cloud service selection tries to support customers in selecting cloud services based on QoS attributes. For considering the right, QoS attributes it is necessary to respect the customers and the providers’ perception. This can be made through a Service Quality Model. Thus, this paper introduces a Cloud Service Quality Model based on a Systematic Literature Review and user interviews as well as providers perceptions.
Download
Paper Nr: 13
Title:

Authentication and Authorization Issues in Mobile Cloud Computing: A Case Study
Authors:

V. Carchiolo, A. Longheu, M. Malgeri, S. Ianniello, M. Marroccia and A. Randazzo
Abstract: Mobile Cloud Computing incorporates Cloud computing paradigma into the mobile environment, the set of technologies that enable to access network services anyplace, anytime and anywhere. This faces many technical challenges, such as low bandwidth, availability, heterogeneity, computing offloads, data accessing, security, privacy, and trust. In this paper the MCC security solution developed and applied within the STMicroelectronics plants is presented.
Download
Paper Nr: 34
Title:

Behind the Surveys: Cloud Adoption at Second Glance
Authors:

Damian Kutzias, Mirjana Stanisic-Petrovic and Claudia Dukino
Abstract: Cloud Computing has evolved from a trend technology to a well established part of the international market and still has growing relevance. For strategic decisions especially of information technology providers as well as governments, surveys can provide relevant information, but as usual in surveys, there are noteworthy differences even for the simplest questions. In this paper we give an overview of several existing cloud surveys and compare some of the questions, particularly related to cloud adoption and scepticism. Differences are highlighted and a list of influencing factors as possible reasons is derived, each with some background, reference proofs and explanation.
Download
Paper Nr: 42
Title:

The Experiential Heterogeneous Earliest Finish Time Algorithm for Task Scheduling in Clouds
Authors:

Artan Mazrekaj, Arlinda Sheholli, Dorian Minarolli and Bernd Freisleben
Abstract: Task scheduling in cloud environments is the problem of assigning and executing computational tasks on the available cloud resources. Effective task scheduling approaches reduce the task completion time, increase the efficiency of resource utilization, and improve the quality of service and the overall performance of the system. In this paper, we present a novel task scheduling algorithm for cloud environments based on the Heterogeneous Earliest Finish Time (HEFT) algorithm, called experiential HEFT. It considers experiences with previous executions of tasks to determine the workload of resources. To realize the experiential HEFT algorithm, we propose a novel way of HEFT rank calculation to specify the minimum average execution time of previous runs of a task on all relevant resources. Experimental results indicate that the proposed experiential HEFT algorithm performs better than HEFT and the popular Critical-Path-on-a-Processor (CPOP) algorithm considered in our comparison.
Download
Paper Nr: 51
Title:

Performance and Cost Analysis between Elasticity Strategies over Pipeline-structured Applications
Authors:

Vinícius Meyer, Miguel G. Xavier, Dionatra F. Kirchoff, Rodrigo R. Righi and Cesar A. F. De Rose
Abstract: With the advances in eScience-related areas and the growing complexity of scientific analysis, more and more scientists are interested in workflow systems. There is a class of scientific workflows that has become a standard to stream processing, called pipeline-structured application. Due to the amount of data these applications need to process nowadays, Cloud Computing has been explored in order to accelerate such processing. However, applying elasticity over stage-dependent applications is not a trivial task since there are some issues that must be taken into consideration, such as the workload proportionality among stages and ensuring the processing flow. There are studies which explore elasticity on pipeline-structured applications but none of them compare or adopt different strategies. In this paper, we present a comparison between two elasticity approaches which consider not only CPU load but also workload processing time information to reorganize resources. We have conducted a number of experiments in order to evaluate the performance gain and cost reduction when applied our strategies. As results, we have reached an average of 72% in performance gain and 73% in cost reduction when comparing non-elastic and elastic executions.
Download
Paper Nr: 55
Title:

Transparent Cloud Privacy: Data Provenance Expression in Blockchain
Authors:

Gabriel Hogan and Markus Helfert
Abstract: The development of Cloud processing and ‘Big Data’ have raised many concerns over the use to which data is being put. These concerns have created new demands for methodologies, and capabilities which can provide transparency and trust in data provenance in the Cloud. Distributed ledger technologies (DLTs) have been proposed as a possible platform to address cloud big data provenance. This paper examines the W3C recommendation for data provenance PROV and if the blockchain DLT can apply the core primary PROV attributes required to satisfy data provenance. The research shows that not all data provenance expressions can be provided by blockchain. Instances of data provenance which rely on circular references are not possible as the blockchain DLT is a single linked list.
Download

Area 2 - Cloud Operations

Short Papers
Paper Nr: 21
Title:

Build-Ship-Run Approach for a CORD-in-a-Box Deployment
Authors:

Ferran Canellas, Nestor Bonjorn, Angelos Mimidis and Jose Soler
Abstract: 5G is expected to provide high bandwidth and low latency communications, thus allowing Telco operators to provide new services to their end customers. This increase in performance is achieved through the migration of network functions from the core to the edge of the network and facilitated by the flexibility and automation provided by Software Defined Networking (SDN) and Network Function Virtualization (NFV). To pave the way to 5G, and simplify the management of 5G deployments a number of SDN/VNF platforms has been developed in the recent years. However deploying and configuring the platforms themselves, is a complex and time consuming task which can act as a barrier to their adoption by Telco operators. This is because Telco Operators strife for fast provisioning times and zero-touch provisioning. Based on this observation, this paper proposes a Build-Ship-Run platform deployment using Central Office Re-architected as a Datacenter (CORD) as an exemplar platform. The proposed approach is based on the use of compressed Virtual Machine snapshots, which allow preconfigured CORD-flavors to be fetched, uncompressed and deployed on demand. Using the proposed workflow, a deployment time seven times better than the raw installation is demonstrated.
Download
Paper Nr: 49
Title:

Policy-based Deployment in a Hybrid and Multicloud Environment
Authors:

Giuseppe Di Modica, Orazio Tomarchio, Hao Wei and Joaquin S. Rodriguez
Abstract: Hybrid and multi-cloud become prominent infrastructure strategy of enterprise. However, the complexity of such infrastructure is a considerable challenge. It is common to see that multi-cloud infrastructure is divided into several smaller units to facilitate the management. The division criteria are geolocation, cost, security, etc. Therefore, how to manage application deployment in such partitioned environment is an intriguing topic of multi-cloud management. We propose a policy-based deployment in multi-cloud infrastructure, which contains policy evaluation and TOSCA standard based orchestration. The system architecture is introduced and a case study with two empirical scenarios is discussed. The results indicate that the proposed policy-based deployment is useful in finding suitable resource and improving deployment efficiency.
Download

Area 3 - Data as a Service

Short Papers
Paper Nr: 74
Title:

Providing Malaria Analytics as a Service
Authors:

Marcos Barreto, Juracy Bertoldo, Alberto Sironi and Vanderson Sampaio
Abstract: Malaria is still a worrying disease worldwide, being responsible for around 219 million cases reported in 2017 and around 435,000 deaths a year. The consensus among researchers, governmental bodies and health professionals is that many countries have relapsed their investments and surveillance actions after a few years of apparent disease reduction. Brazil is within such countries and, consequently, is presenting a constant increase in the number of reported cases since 2016 (more than 20% a year). Given this context, the National Malaria Control Program (NMCP) promotes several actions to redirect the country towards the malaria elimination path. Among such actions, the improvement of the surveillance ecosystem is considered crucial to allow efficacy of control actions, including vector control as well as early diagnosis and prompt treatment. In this paper, we present our efforts in designing a visual mining tool allowing descriptive and predictive analytics over an integrated database comprising malaria surveillance data, climate and vector control data. This tool has been used as a “data service” by NMCP and partner researchers for validation purposes. So far, our results have demonstrated that surveillance and combat actions can be highly improved by using this tool.
Download

Area 4 - Edge Cloud and Fog Computing

Full Papers
Paper Nr: 23
Title:

A Containerized Big Data Streaming Architecture for Edge Cloud Computing on Clustered Single-board Devices
Authors:

Remo Scolati, Ilenia Fronza, Nabil El Ioini, Areeg Samir and Claus Pahl
Abstract: The constant increase of the amount of data generated by Internet of Things (IoT) devices creates challenges for the supporting cloud infrastructure, which is often used to process and store the data. This work focuses on an alternative approach, based on the edge cloud computing model, i.e., processing and filtering data before transferring it to a backing cloud infrastructure. We describe the implementation of a low-power and low-cost cluster of single board computers (SBC) for this context, applying models and technologies from the Big Data domain with the aim of reducing the amount of data which has to be transferred elsewhere. To implement the system, a cluster of Raspberry Pis was built, relying on Docker to containerize and deploy an Apache Hadoop and Apache Spark cluster, on which a test application is then executed. A monitoring stack based on Prometheus, a popular monitoring and alerting tool in the cloud-native industry, is used to gather system metrics and analyze the performance of the setup. We evaluate the complexity of the system, showing that by means of containerization increased fault tolerance and ease of maintenance can be achieved, which makes the proposed solution suitable for an industrial environment. Furthermore, an analysis of the overall performance, which takes into account the resource usage of the proposed solution with regards to the constraints imposed by the devices, is presented in order to discuss the capabilities and limitations of proposed architecture.
Download
Paper Nr: 24
Title:

A Fog Computing Service Placement for Smart Cities based on Genetic Algorithms
Authors:

Claudia Canali and Riccardo Lancellotti
Abstract: The growing popularity of the Fog Computing paradigm is driven by the increasing availability of large amount of sensors and smart devices on a geographically distributed area. The scenario of a smart city is a clear example of this trend. As we face an increasing presence of sensors producing a huge volume of data, the classical cloud paradigm, with few powerful data centers that are far away from the data sources, becomes inadequate. There is the need to deploy a highly distributed layer of data processors that filter, aggregate and pre-process the incoming data according to a fog computing paradigm. However, a fog computing architecture must distribute the incoming workload over the fog nodes to minimize communication latency while avoiding overload. In the present paper we tackle this problem in a twofold way. First, we propose a formal model for the problem of mapping the data sources over the fog nodes. The proposed optimization problem considers both the communication latency and the processing time on the fog nodes (that depends on the node load). Furthermore, we propose a heuristic, based on genetic algorithms to solve the problem in a scalable way. We evaluate our proposal on a geographic testbed that represents a smart-city scenario. Our experiments demonstrate that the proposed heuristic can be used for the optimization in the considered scenario. Furthermore, we perform a sensitivity analysis on the main heuristic parameters.
Download
Paper Nr: 26
Title:

FUSE: A Microservice Approach to Cross-domain Federation using Docker Containers
Authors:

Tom Goethals, Dwight Kerkhove, Laurens Van Hoye, Merlijn Sebrechts, Filip De Turck and Bruno Volckaert
Abstract: In crisis situations, it is important to be able to quickly gather information from various sources to form a complete and accurate picture of the situation. However, the different policies of participating companies often make it difficult to connect their information sources quickly, or to allow software to be deployed on their networks in a uniform way. The difficulty in deploying software is exacerbated by the fact that companies often use different software platforms in their existing networks. In this paper, Flexible federated Unified Service Environment (FUSE) is presented as a solution for joining multiple domains into a microservice based ad hoc federation, and for deploying and managing container-based software on the devices of a federation. The resource requirements for setting up a FUSE federation are examined, and a video streaming application is deployed to demonstrate the performance of software deployed on an example federation. The results show that FUSE can be deployed in 10 minutes or less, and that it can support multiple video streams under normal network conditions, making it a viable solution for the problem of quick and easy cross-domain federation.
Download
Short Papers
Paper Nr: 56
Title:

GaruaGeo: Global Scale Data Aggregation in Hybrid Edge and Cloud Computing Environments
Authors:

Otávio Carvalho, Eduardo Roloff and Philippe A. Navaux
Abstract: The combination of Edge Computing devices and Cloud Computing resources brings the best of both worlds: Data aggregation closer to the source and scalable resources to grow the network on demand. However, the ability to leverage each time more powerful edge nodes to decentralize data processing and aggregation is still a significant challenge for both industry and academia. In this work, we extend the Garua platform to analyze the impact of a model for data aggregation in a global scale smart grid application dataset. The platform is extended to support global data aggregators that are placed nearly to the Edge nodes where data is being collected. This way, it is possible to aggregate data not only at the edge of the network but also pre-process data at nearby geographic areas, before sending data to be aggregated globally by global centralization nodes. The results of this work show that the implemented testbed application, through the usage of edge node aggregation, data aggregators geographically distributed and messaging windows, can achieve collection rates above 400 million measurements per second.
Download
Paper Nr: 89
Title:

Situation-Aware Management of Cyber-Physical Systems
Authors:

Kálmán Képes, Uwe Breitenbücher and Frank Leymann
Abstract: The current trend of connecting the physical world with the so-called cyber world resulted in paradigms such as the Internet of Things or the more general paradigm of Cyber-Physical Systems. The wide range of domains applicable results in a heterogeneous landscape of software and hardware solutions. To benefit of the paradigm, developers must be able to integrate different solutions from a range of different domains. However, these systems must therefore be able to change components, configurations and environments, hence, be adaptable at runtime. We present an approach that is based on the combination of Situation-Aware Adaptation concepts and Deployment Models. The general idea is to start processes that can change application structure and configuration when a certain situation in the context of applications occur. We validated the technical feasibility of our approach by a prototypical implementation based on a Smart Home scenario.
Download
Paper Nr: 90
Title:

A Containerized Tool to Deploy Scientific Applications over SoC-based Systems: The Case of Meteorological Forecasting with WRF
Authors:

Luiz A. Steffenel, Andrea S. Charão and Bruno S. Alves
Abstract: Container-based virtualization represents a flexible and scalable solution for HPC environments, allowing a simple and efficient management of scientific applications. Recently, Systems-on-a-Chip (SoC) have emerged as an alternative to traditional HPC clusters, with a good computing power and low costs. In this paper, we present how we developed a container-based solution for SoC clusters, and study the performance of WRF (Weather Research and Forecasting) in such environments. The results demonstrate that although the peak performance of SoC clusters is still limited, these environments are more than suitable to scientific application that have relaxed QoS constraints.
Download
Paper Nr: 4
Title:

A Fog-Cloud Computing Infrastructure for Condition Monitoring and Distributing Industry 4.0 Services
Authors:

Timo Bayer, Lothar Moedel and Christoph Reich
Abstract: Data-driven Industry 4.0 applications require low latency data processing and reliable communication models to enable efficient operation of production lines, fast response to failures and quickly adapt the manufacturing process to changing environmental conditions. Data processing in the Cloud has been widely accepted and in combination with Fog technologies, it can also satisfy these requirements. This paper investigates the placement of service containers and wheater they should be carried out in the Cloud or at a Fog node. It shows how to provide an uniform well-monitored execution environment to automatically distribute services concerning their application-specific requirements. An infrastructure is presented, that utilizes measurement probes to observe the node and environmental conditions, derive and evaluate appropriate distribution algorithms and finally deploy the application services to the node that meets the requirements.
Download

Area 5 - Mobile Cloud Computing

Short Papers
Paper Nr: 39
Title:

Improving the Trade-Off between Performance and Energy Saving in Mobile Devices through a Transparent Code Offloading Technique
Authors:

Rômulo Reis, Paulo Souza, Wagner Marques, Tiago Ferreto and Fábio D. Rossi
Abstract: The popularity of mobile devices has increased significantly, and nowadays they are used for the most diverse purposes like accessing the Internet or helping on business matters. Such popularity emerged as a consequence of the compatibility of these devices with a large variety of applications. However, the complexity of these applications boosted the demand for computational resources on mobile devices. Code Offloading is a solution that aims to mitigate this problem by reducing the use of resources and battery on mobile devices by sending parts of applications to be processed in the cloud. In this sense, this paper presents an evaluation of a transparent code offloading technique, where no modification in the application source code is required to allow the smartphone to send parts of the application to be processed in the cloud. We used a face detection application for the evaluation. Results showed the technique can improve applications performance in some scenarios, achieving speed-up of 12x in the best case.
Download

Area 6 - Service Modelling and Analytics

Full Papers
Paper Nr: 64
Title:

Cost-evaluation of Cloud Portfolios: An Empirical Case Study
Authors:

Benedikt Pittl, Werner Mach and Erich Schikuta
Abstract: Today, Amazon is the Cloud service market leader with the EC2 platform. Three predominant marketspaces exist on this platform: spot marketspace, reservation marketspace, and the well-known on-demand marketspace. Also other providers such as Google, Microsoft and VirtuStream run multiple marketspaces. Consumers can purchase their virtual machines from different providers on different marketspace to form Cloud portfolios: a bundle of virtual machines whereby the virtual machines have different technical characteristics and pricing mechanisms. An industry-relevant research challenge is to provide best practices and guidelines for creating cost-efficient Cloud portfolios. In this paper, we used Amazon’s marketspaces and the dataset from the Bitbrains datacenter to analyze the cost-efficiency of heterogeneous Cloud portfolios - portfolios where the virtual machines are purchased from different marketspaces. We found out that heterogeneous portfolios are more cost-efficient than homogeneous portfolios for almost all analyzed situations. Our analysis further revealed that consumers request virtual machines that are over-sized which forms a significant field of cost-optimization. A second dataset from the Bitbrains datacenter - from another domain of application - validates our findings.
Download
Short Papers
Paper Nr: 27
Title:

Towards the Modelling of Adaptation Rules and Histories for Multi-Cloud Applications
Authors:

Kyriakos Kritikos, Chrysostomos Zeginis, Eleni Politaki and Dimitris Plexousakis
Abstract: Currently, there is a move towards adopting multi-clouds due to their main benefits, including vendor lock-in avoidance and optimal application realisation via different cloud services. However, such multi-cloud applications face a new challenge related to the dynamicity and uncertainty that even a single cloud environment exhibits. As such, they cannot deliver a suitable service level to their customers, resulting in SLA penalty costs and application provider reputation reduction. To this end, we have previously proposed a cross-level and multi-cloud application adaptation architecture. Towards realising this architecture, this paper proposes two extensions of the CAMEL language allowing to specify advanced adaptation rules and histories. Such extensions not only enable to cover cross-level application adaptation by executing adaptation workflows but also to progress such an adaptation to address both the application and exploited cloud services evolution.
Download
Paper Nr: 72
Title:

An Approach to Determine & Apply Solutions to Solve Detected Problems in Restructured Deployment Models using First-order Logic
Authors:

Karoline Saatkamp, Uwe Breitenbücher, Michael Falkenthal, Lukas Harzenetter and Frank Leymann
Abstract: New paradigms such as edge computing opened up new opportunities for distributing applications to meet use-case-specific requirements. For automating the deployment of applications, deployment models can be created that describe the application structure with its components and their relations. However, the distribution is often not known in advance and, thus, deployment models have to be restructured. This can result in problems that have not existed before, e.g., components previously deployed in the same network were distributed, but security mechanisms are missing. Architecture patterns can be used to detect such problems, however, patterns describe only generic technology-independent solutions, which cannot automatically be applied to applications. Several concrete technologies exist that implements the pattern. Which solutions are applicable to a particular application is determined by, e.g., its hosting environment or used communication protocol. However, the manual effort to determine and implement appropriate solutions is immense. In this work, we present an approach to automate (i) the determination of solutions for an application using first-order logic and (ii) the adaptation of its deployment model accordingly. To validate the practical feasibility, we present a prototype using the cloud standard TOSCA and the logic programming language PROLOG.
Download
Paper Nr: 68
Title:

Function-as-a-Service Benchmarking Framework
Authors:

Roland Pellegrini, Igor Ivkic and Markus Tauber
Abstract: Cloud Service Providers deliver their products in form of ”as-a-Service”, which are typically categorized by the level of abstraction. This approach hides the implementation details and shows only functionality to the user. However, the problem is that it is hard to measure the performance of Cloud services, because they behave like black boxes. Especially with Function-as-a-Service it is even more difficult because it completely hides server and infrastructure management from users by design. Cloud Service Prodivers usually restrict the maximum size of code, memory and runtime of Cloud Functions. Nevertheless, users need clarification if more ressources are needed to deliver services in high quality. In this regard, we present the architectural design of a new Function-as-a-Service benchmarking tool, which allows users to evaluate the performance of Cloud Functions. Furthermore, the capabilities of the framework are tested on an isolated platform with a specific workload. The results show that users are able to get insights into Function-as-a-Service environments. This, in turn, allows users to identify factors which may slow down or speed up the performance of Cloud Functions.
Download
Paper Nr: 71
Title:

A Framework for Measuring the Costs of Security at Runtime
Authors:

Igor Ivkic, Harald Pichler, Mario Zsilak, Andreas Mauthe and Markus Tauber
Abstract: In Industry 4.0, Cyber-Physical Systems (CPS) are formed by components, which are interconnected with each other over the Internet of Things (IoT). The resulting capabilities of sensing and affecting the physical world offer a vast range of opportunities, yet, at the same time pose new security challenges. To address these challenges there are various IoT Frameworks, which offer solutions for managing and controlling IoT-components and their interactions. In this regard, providing security for an interaction usually requires performing additional security-related tasks (e.g. authorisation, encryption, etc.) to prevent possible security risks. Research currently focuses more on designing and developing these frameworks and does not satisfactorily provide methodologies for evaluating the resulting costs of providing security. In this paper we propose an initial approach for measuring the resulting costs of providing security for interacting IoT-components by using a Security Cost Modelling Framework. Furthermore, we describe the necessary building blocks of the framework and provide an experimental design showing how it could be used to measure security costs at runtime.
Download

Area 7 - Services Science

Full Papers
Paper Nr: 9
Title:

Towards a Roadmap for the Migration of Legacy Software Systems to a Microservice based Architecture
Authors:

Hugo O. S. da Silva, Glauco F. Carneiro and Miguel P. Monteiro
Abstract: The migration of legacy software systems to a microservice based architecture is not a trivial task due to challenges and difficulties as reported in the literature. The concept of microservices mainly consists in software organized as a suite of small, modular, and independently deployed services that run on their own processes and communicate through well-defined, lightweight mechanisms to serve a business goal. However, the literature is still incipient in relation to step-by-step guidelines supporting practitioners to accomplish the migration from an existing, monolithic structure to a microservice based architecture. Goal: Discuss lessons learned from the migration of legacy software systems to microservices-based architecture. Method: We conducted two studies (a pilot and a case study) aiming at characterizing the relevants steps of such guidelines. Results: We report the steps and challenges observed during the migration reported in this study. Conclusion: We identify at least three main phases that drive the migration process.
Download
Paper Nr: 66
Title:

From Monolithic Systems to Microservices: A Decomposition Framework based on Process Mining
Authors:

Davide Taibi and Kari Systä
Abstract: Decomposition is one of the most complex tasks during the migration from monolithic systems to microservices, generally performed manually, based on the experience of the software architects. In this work, we propose a 6-step framework to reduce the subjectivity of the decomposition process. The framework provides software architects with a set of decomposition options, together with a set of measures to evaluate and compare their quality. The decomposition options are identified based on the independent execution traces of the system by means of the application of a process-mining tool to the log traces collected at runtime. We validated the process, in an industrial project, by comparing the proposed decomposition options with the one proposed by the software architect that manually analyzed the system. The application of our framework allowed the company to identify issues in their software that the architect did not spot manually, and to discover more suitable decomposition options that the architect did not consider. The framework could be very useful also in other companies to improve the quality of the decomposition of any monolithic system, identifying different decomposition strategies and reducing the subjectivity of the decomposition process. Moreover, researchers could extend our approach increasing the support and further automating the decomposition support.
Download
Short Papers
Paper Nr: 3
Title:

A Backup-as-a-Service (BaaS) Software Solution
Authors:

Heitor Faria, Priscila Solís, Jarcir Bordim and Rodrigo Hagstrom
Abstract: Backup is a replica of any data that can be used to restore its original form. However, the total amount of digital data created worldwide more than doubles every two years and is expected to reach 44 trillions of gigabytes in 2020, bringing constant new challenges to backup processes. Enterprise backup is one of the oldest and most performed tasks by infrastructure and operations professionals. Still, most backup systems have been designed and optimized for outdated environments and use cases. That fact, generates frustration over currently backup challenges and leads to a greater willingness to modernize and to consider new technologies. Traditional backup and archive solutions are no longer able to meet users current needs. The ideal modern backup and recovery software should not only provide features to attend a traditional data center, but also allow the integration and exploration of the growing Cloud, including “backup client as a service” and “backup storage as a service”. The present study proposed and deploys a Backup as a Service software solution. For that, the cloud/backup parameters, cloud backup challenges, researched architectures and Backup-as-a-Service (BaaS) system requirements are specified. Then, a selected set of BaaS desired features are developed, resulting in the first truly cloud REST API based Backup-as-a-Service interface, namely “bcloud”. Finally, this work conducts an on-line usability poll with a significant number of users. The analysis of results in an overall average objective zero to ten questions evaluation was 8.29%, indicating a very satisfactory user perception of the bcloud BaaS interface prototype.
Download
Paper Nr: 16
Title:

Towards Early Prototyping of Services based on Open Transport Data: A Feasibility Study
Authors:

Nicolas Ferry, Aida Omerovic and Marit K. Natvig
Abstract: Data under open licenses and in reusable formats, often referred to as ”open data”, is increasingly being made accessible by both public and private actors. Government institutions, municipalities, private companies and entrepreneurs are among the stakeholders either having visions of new open data-based services, or just looking for new ideas on potential innovations based on open data. It is, however, in both cases, often unclear to the service developers how the open data actually can be utilized. A main reason is that the data needs to be retrieved from multiple sources, understood, quality checked and processed. While gaining insights on possible services that can be created on the top of open data, a service developer has to undergo an iterative ”trying and failing” exercise of service prototyping. In order to be practically feasible, such a process needs to be agile and efficient. Open data from the transport sector is used as a case. The open transport data are characterized by many challenges common for open data in general, but also a few specific ones. One of those challenges is the need for combining (often real-time) data from rather many sources in order to create a new service. In this paper we propose an agile approach to early service prototyping and we try out the approach on an open transport data service. Finally, we propose the priorities for future work towards a comprehensive approach for agile prototyping of open transport data-based services.
Download
Paper Nr: 22
Title:

On the Impacts of Transitive Indirect Reciprocity on P2P Cloud Federations
Authors:

Eduardo L. Falcão, Antônio A. Neto, Francisco Brasileiro and Andrey Brito
Abstract: Several P2P systems of resource sharing use cooperation incentive mechanisms to identify and punish free riders, i.e., non-reciprocal individuals. A widespread approach is to use the levels of reciprocity, either directly or indirectly, to decide the extent to which an individual should trust other partners. One restriction of direct reciprocity mechanisms is the inability to foster cooperation between individuals with asymmetrical resources or availability incompatibility. In this work, we evaluate the performance of cloud federations ruled by the combination of the well-known direct reciprocity with transitive reciprocity, a strategy that allows direct reciprocity mechanisms to deal with asymmetry between individuals, while still keeping the benefits of direct reciprocity. For this, we implemented a simulator of resource bartering in cloud federations and experimented it with workloads synthesized from traces of real systems. Our best results showed an average increase of 12.83% and 26.38% on the sharing level of the federation, in an optimistic but unrealistic mechanism setup. When configured in a feasible and realistic manner, the transitive reciprocity was able to increase the sharing level up to an average of 6.02% and 7.53%.
Download
Paper Nr: 36
Title:

Service-oriented Mogramming with SML and SORCER
Authors:

Michael Sobolewski
Abstract: Service-oriented Mogramming Language (SML) is designed for service-orientation as UML was considered for object-orientation. SML is an executable language in the SORCER platform based on service abstraction (everything is a service) and three pillars of service-orientation: context awareness (contexting), multifidelity, and multityping. Context awareness is related to parametric polymorphism, multifidelity is related to ad hoc polymorphism, and multityping is a form of net-centric type polymorphism. SML allows for defining polymorphic service systems that can reconfigure and morph service federations at runtime. In this paper the basic concepts of SML are presented with three ted design patterns of service federations. Its runtime environment is introduced with the focus on the presented service abstractions.
Download
Paper Nr: 40
Title:

An Ontological Framework for Reasoning about Relations between Complex Access Control Policies in Cloud Environments
Authors:

Simeon Veloudis, Iraklis Paraskakis and Christos Petsos
Abstract: By embracing the cloud computing paradigm enterprises are able to realise significant cost savings whilst boosting their agility and productivity. Yet, due mainly to security and privacy concerns, many enterprises are reluctant to migrate the storage and processing of their critical assets to the cloud. One way to alleviate these concerns, hence bolster the adoption of cloud computing, is to infuse suitable access control policies in cloud services. Nevertheless, the complexity inherent in such policies, stemming from the dynamic nature of cloud environments, calls for a framework capable of providing assurances with respect to the effectiveness of these policies. The work presented in this paper elaborates on such a framework. In particular, it proposes an approach for generically checking potential subsumption relations between access control policies that incorporate the contextual knowledge that characterises an access request and which needs to be taken into account for granting, or denying, the request. The proposed framework is expressed ontologically hence enabling automated reasoning, through semantic inferencing, about policy subsumption.
Download
Paper Nr: 92
Title:

DRAGON: A Dynamic Scheduling and Scaling Controller for Managing Distributed Deep Learning Jobs in Kubernetes Cluster
Authors:

Chan-Yi Lin, Ting-An Yeh and Jerry Chou
Abstract: With the fast growing trend in deep learning driven AI services over the past decade, deep learning, especially the resource-intensive and time-consuming training jobs, have become one of the main workload in today’s production clusters. However, due to the complex workload characteristics of deep learning, and the dynamic natural of shared resource environment, managing the resource allocation and execution lifecycle of distributed training jobs in cluster can be challenging. This work aims to address these issues by developing and implementing a scheduling and scaling controller to dynamically manage distributed training jobs on a Kubernetes (K8S) cluster, which is a broadly used platform for managing containerized workloads and services. The objectives of our proposed approach is to enhance K8S with three capabilities: (1) Task dependency aware gang scheduling to avoid idle resources. (2) Locality aware task placement to minimize communication overhead. (3) Load aware job scaling to improve cost efficiency. Our approach is evaluated by real testbed and simulator using a set of TensorFlow jobs. Comparing to the default K8S scheduler, our approach successfully improved resource utilization by 20% ∼ 30% and reduced job elapsed time by over 65%.
Download
Paper Nr: 31
Title:

The Elastic Processing of Data Streams in Cloud Environments: A Systematic Mapping Study
Authors:

Floriment Klinaku, Michael Zigldrum, Markus Frank and Steffen Becker
Abstract: Ongoing efforts exist to exploit cloud elasticity for processing efficiently data streams generated by a variety of data sources. To contribute to these efforts an overview of existing research is required. To the best of our knowledge, a systematic overview of the field is missing. To fill this gap, we conduct a Systematic Literature Map (SLM). This way we offer a high-level overview of the literature on elastic data stream processing. We search four databases, evaluate 564 publications and identify 100 relevant publications. The identified publications show that the majority of work is validated research through proofs-of-concept and very few through case studies, surveys and field experiments. There are several frameworks, approaches and tools proposed, but, fewer metrics, models and processes.
Download
Paper Nr: 73
Title:

A Comparison of Multi-cloud Provisioning Platforms
Authors:

Domenico Calcaterra, Vincenzo Cartelli, Giuseppe Di Modica and Orazio Tomarchio
Abstract: Although cloud computing has been around for over a decade now, many issues of the first hour still persist. Vendor lock-in, poor interoperability and portability hinder users from taking full advantage of main cloud features. On the industry side, the big players often adopt proprietary solutions to guarantee a seamless management and orchestration of cloud applications; on the research side, the TOSCA specification has emerged as the most authoritative effort for the interoperable description of cloud services. In this paper, we discuss the design and implementation of a TOSCA-based orchestration framework for the automated service provisioning. We also compare this approach to two open-source orchestration tools (Heat, Cloudify) with respect to the deployment of a two-tier application on an OpenStack environment. Test results show our method performs comparably to the aforementioned tools, while maintaining full compliance with TOSCA.
Download

Area 8 - Cloud Computing Platforms and Applications

Full Papers
Paper Nr: 7
Title:

Unifying Data and Replica Placement for Data-intensive Services in Geographically Distributed Clouds
Authors:

Ankita Atrey, Gregory Van Seghbroeck, Higinio Mora, Filip De Turck and Bruno Volckaert
Abstract: The increased reliance of data management applications on cloud computing technologies has rendered research in identifying solutions to the data placement problem to be of paramount importance. The objective of the classical data placement problem is to optimally partition, while also allowing for replication, the set of data-items into distributed data centers to minimize the overall network communication cost. Despite significant advancement in data placement research, replica placement has seldom been studied in unison with data placement. More specifically, most of the existing solutions employ a two-phase approach: 1) data placement, followed by 2) replication. Replication should however be seen as an integral part of data placement, and should be studied as a joint optimization problem with the latter. In this paper, we propose a unified paradigm of data placement, called CPR, which combines data placement and replication of data-intensive services into geographically distributed clouds as a joint optimization problem. Underneath CPR, lies an overlapping correlation clustering algorithm capable of assigning a data-item to multiple data centers, thereby enabling us to jointly solve data placement and replication. Experiments on a real-world trace-based online social network dataset show that CPR is effective and scalable. Empirically, it is  35% better in efficacy on the evaluated metrics, while being up to 8 times faster in execution time when compared to state-of-the-art techniques.
Download
Paper Nr: 45
Title:

Handling Packet Losses in Cloud-based MPTCP Application Traffic
Authors:

Kiran Yedugundla, Per Hurtig and Anna Brunstrom
Abstract: Internet traffic is comprised of data flows from various applications with unique traffic characteristics. For many cloud applications, end-to-end latency is a primary factor affecting the perceived user experience. As packet losses cause delays in the communication they impact user experience, making efficient handling of packet losses an important function of transport layer protocols. Multipath TCP (MPTCP) is a modification to TCP that enables simultaneous use of several paths for a TCP flow. MPTCP is known to improve throughput. However, the performance of MPTCP is not optimal when handling certain loss scenarios. Efficient packet loss recovery is thus important to achieve desirable flow completion times for interactive cloud-based applications. In this paper we evaluate the performance of MPTCP in handling tail losses using traffic traces from various cloud-based applications. Tail losses, losses that occur at the end of a flow or traffic burst, are particularly challenging from a latency perspective as they are difficult to detect and recover in a timely manner. Tail losses in TCP are handled by using a tail loss probe (TLP) mechanism which was adapted to MPTCP from TCP. We investigate the performance of TLP in MPTCP, comparing the standard implementation to a recently proposed, less conservative approach. Our experimental results show that a less conservative implementation of TLP performs significantly better than the standard implementation in handling tail losses, reducing the average burst completion time of cloud based applications when tail loss occurs by up to 50% in certain cases.
Download
Paper Nr: 47
Title:

Human-Computer Cloud: Application Platform and Dynamic Decision Support
Authors:

A. Smirnov, N. Shilov, A. Ponomarev and M. Schekotov
Abstract: The paper describes a human-computer cloud environment supporting the deployment and functioning of human-based applications and allowing to decouple computing resource management issues (for this kind of applications) from application software. The paper focuses on two specific contributions lying in the heart of the proposed human-computer cloud environment: a) application platform, allowing to deploy human-based applications and using digital contracts to regulate the interactions between an application and its contributors, and b) the principles of ontology-based decision support service that is implemented on top of the human-computer cloud and uses task decomposition in order to deal with ad hoc tasks, algorithms for which are not described in advance.
Download
Paper Nr: 65
Title:

IoLT Smart Pot: An IoT-Cloud Solution for Monitoring Plant Growth in Greenhouses
Authors:

J. Hadabas, M. Hovari, I. Vass and A. Kertesz
Abstract: According to a recent Beecham Research report, food production have to be increased by 70 percent till 2050 to feed 9.6 billion global population predicted by the United Nations Food and Agriculture Organisation. Since Cloud Computing and the Internet of Things (IoT) have already opened new ways for revolutionizing industrial processes, these technologies could be important for the farming industry. Smart farming has the potential to improve productivity and reduce waste to transform agriculture. Plant phenotyping is an important research field that gained a high attention recently due to the need for complex monitoring of development and stress responses of plants. However, the current phenotyping platforms are very expensive, and used in large central infrastructures, which limit their widepread use. The newly emerging ICT technologies together with the availability of low cost sensors and computing solutions paved the way towards the development of affordable phenotyping solutions, which can be applied under standard greenhouse conditions. The Internet of Living Things (IoLT) project has been launched to integrate IoT technological research with applied research on specific, biological applications. In this paper we introduce our research results for developing a low cost plant phenotyping platform for small sized plants, which is one of our goals in this project. The proposed IoLT Smart Pot is capable of monitoring environmental parameters by sensors placed above the plant and into the pot, managed by a Raspberry Pi board placed under the pot. We have also developed a private IoT-Cloud gateway for receiving, storing, visualizing and downloading the monitored parameters sent by the pot devices. We have performed the evaluation of our proposed platform both with simulated and real smart pots.
Download
Paper Nr: 80
Title:

ArchaDIA: An Architecture for Big Data as a Service in Private Cloud
Authors:

Marco S. Reis and Aletéia P. Favacho de Araújo
Abstract: There are multiple definitions and technologies making the path to a big data solution a challenging task. The use of cloud computing together with a proven big data software architecture helps reducing project costs, development time and abstracts the complexity of the underlying implementation technologies. The combination of cloud computing and big data platforms results in a new service model, called Big Data as a Service (BDaaS), that automates the process of provisioning the infrastructure. This paper presents an architecture for big data systems in private clouds, using a real system to evaluate the functionalities. The architecture supports batch/real-time processing, messaging systems and data services based on web APIs. The architectural description defines the technology roadmap, composed exclusively of big data tools. The results showed that the proposed architecture supports the facilities of cloud computing and performs well in the analysis of large datasets.
Download
Paper Nr: 82
Title:

TASKWORK: A Cloud-aware Runtime System for Elastic Task-parallel HPC Applications
Authors:

Stefan Kehrer and Wolfgang Blochinger
Abstract: With the capability of employing virtually unlimited compute resources, the cloud evolved into an attractive execution environment for applications from the High Performance Computing (HPC) domain. By means of elastic scaling, compute resources can be provisioned and decommissioned at runtime. This gives rise to a new concept in HPC: Elasticity of parallel computations. However, it is still an open research question to which extent HPC applications can benefit from elastic scaling and how to leverage elasticity of parallel computations. In this paper, we discuss how to address these challenges for HPC applications with dynamic task parallelism and present TASKWORK, a cloud-aware runtime system based on our findings. TASKWORK enables the implementation of elastic HPC applications by means of higher-level development frameworks and solves corresponding coordination problems based on Apache ZooKeeper. For evaluation purposes, we discuss a development framework for parallel branch-and-bound based on TASKWORK, show how to implement an elastic HPC application, and report on measurements with respect to parallel efficiency and elastic scaling.
Download
Paper Nr: 88
Title:

Exploring Instance Heterogeneity in Public Cloud Providers for HPC Applications
Authors:

Eduardo Roloff, Matthias Diener, Luciano P. Gaspary and Philippe A. Navaux
Abstract: Public cloud providers offer a wide range of instance types with different speeds, configurations, and prices, which allows users to choose the most appropriate configurations for their applications. When executing parallel applications that require multiple instances to execute, such as large scientific applications, most users pick an instance type that fits their overall needs best, and then create a cluster of interconnected instances of the same type. However, the tasks of a parallel application often have different demands in terms of performance and memory usage. This difference in demands can be exploited by selecting multiple instance types that are adapted to the demands of the application. This way, the combination of public cloud heterogeneity and application heterogeneity can be exploited in order to reduce the execution cost without significant performance loss. In this paper we conduct an evaluation of three major public cloud providers: Microsoft, Amazon, and Google, comparing their suitability for heterogeneous execution. Results show that Azure is the most suitable of the three providers, with cost efficiency gains of up to 50% compared to homogeneous execution, while maintaining the same performance.
Download
Short Papers
Paper Nr: 17
Title:

Mapping of Quality of Service Requirements to Resource Demands for IaaS
Authors:

Ioannis Bouras, Fotis Aisopos, John Violos, George Kousiouris, Alexandros Psychas, Theodora Varvarigou, Gerasimos Xydas, Dimitrios Charilas and Yiannis Stavroulas
Abstract: Deciding and reserving appropriate resources in the Cloud, is a basic initial step for adopters when employing an Infrastructure as a Service to host their application. However, the size and number of Virtual Machines used, along with the expected application workload, will highly influence its operation, in terms of the observed Quality of Service. This paper proposes a machine learning approach, based on Artificial Neural Networks, for mapping Quality of Service required levels and (expected) application workload to concrete resource demands. The presented solution is evaluated through a comercial Customer Relationship Management application, generating a training set of realistic workload and Quality of Service measurements in order to illustrate the effectiveness of the proposed technique in a real-world scenario.
Download
Paper Nr: 20
Title:

Performance Prediction of GPU-based Deep Learning Applications
Authors:

Eugenio Gianniti, Li Zhang and Danilo Ardagna
Abstract: Recent years saw an increasing success in the application of deep learning methods across various domains and for tackling different problems, ranging from image recognition and classification to text processing and speech recognition. In this paper we propose, discuss, and validate a black box approach to model the execution time for training convolutional neural networks (CNNs), with a particular focus on deployments on general purpose graphics processing units (GPGPUs). We demonstrate that our approach is generally applicable to a variety of CNN models and different types of GPGPUs with high accuracy. The proposed method can support with great precision (within 5% average percentage error) the management of production environments.
Download
Paper Nr: 41
Title:

SeeDep: Deploying Reproducible Application Topologies on Cloud Platform
Authors:

Cyril Seguin, Eddy Caron and Samuel Dubus
Abstract: As part of the scientific method, any researcher should be able to reproduce the experimentation in order to not only verify the result but also evaluate and compare this experimentation with other approaches. The need of a standard tool allowing researchers to easily generate, share and reproduce experiments set-up arises. In this paper, we present SeeDep, a framework aiming at being such a standard tool. By associating a generation key to a network experiment set-up, SeeDep allows for reproducing network experiments independently from the used infrastructure.
Download
Paper Nr: 50
Title:

IAGREE: Infrastructure-agnostic Resilience Benchmark Tool for Cloud Native Platforms
Authors:

Paulo Souza, Wagner Marques, Rômulo Reis and Tiago Ferreto
Abstract: The cloud native approach is getting more and more popular with the proposal of decomposing application into small components called microservices, which are designed to minimize the costs with upgrades and maintenance and increase the resources usage efficiency. However, microservices architecture brings some challenges such as preserving the manageability of the platforms, since the greater the number of microservices the applications have, the greater the complexity of ensuring that everything is working as expected. In this context, one of the concerns is to evaluate the resilience of platforms. Current resilience benchmark tools are designed for running in specific infrastructures. Thus, in this paper we present IAGREE, a benchmark tool designed for measuring multiple resilience metrics in cloud native platforms based on Cloud Foundry and running upon any infrastructure.
Download
Paper Nr: 52
Title:

Cloud.Jus: Architecture for Provisioning Infrastructure as a Service in the Government Sector
Authors:

Klayton Castro, Gabriel D. Macedo, Aleteia F. Araujo and Leonardo R. de Carvalho
Abstract: Building a community cloud by federating private clouds is one of the lower cost alternatives for hosting applications that require distributed deployment to meet scale-saving, high availability, reliability, and service level compliance. Despite its potential benefits, there are many issues about lack of standardization, system integration, interoperability and portability across multiple service providers, resulting in low adherence to the model in organizations that are still struggling to adapt its legacy applications to a cloud architecture in complex environments, such as some governmental sector scenarios. Currently, there is no seamless approach to migrate from the traditional infrastructure model to a cloud computing model on these organizations. So, we propose an architecture for building a community cloud even in scenarios with a strong presence of non-cloud native applications by developing a low-coupled infrastructure middleware that supports different hypervisors, a GUI and a CLI. To show the feasibility of our approach, we evaluate the architecture on a set of infrastructures at Superior Courts of Brazilian Judicial Branch, that may compose a cost-effective solution to start the transition to the cloud model in other organizations also.
Download
Paper Nr: 57
Title:

Privacy Enhancing Data Access Control for Ambient Assisted Living
Authors:

Hendrik Kuijs, Timo Bayer, Christoph Reich, Martin Knahl and Nathan Clarke
Abstract: As private data is key to applications in the field of Ambient Assisted Living, access control has to be in place to regulate data flows within the environment and to preserver the privacy of a user. We present a data access control system based on an easy to understand policy language with the ability to be extended by context information. Context information are enabling applications in the field of Ambient Assisted Living to adapt their behaviour to temporal, emergency or environmental conditions. The system is able to monitor and control data flows in OSGi environments by proxy services, without the need of modifying the core platform or the service logic of bundles. This is necessary to inform data subjects, to enable the data subject to control the environment and to enforce data access policies that are compatible with legal requirements.
Download
Paper Nr: 76
Title:

Cloud-based Conversational Agents for User Acquisition and Engagement
Authors:

Manasés J. Galindo Bello
Abstract: The benefits of cloud computing have driven different companies from diverse sectors to migrate their products and services to the cloud. In the last decade many businesses have adopted web and mobile applications to offer better customer service as well as used social networks for advertisements and marketing campaigns aiming to acquire, engage and retain their customers. This paper presents a case study combining the areas of chatbots, cloud computing and customer service, acquisition and engagement targeting the gastronomy industry; it evaluates and compares the implementation of a chatbot as a cloud-native application (Platform as a Service) versus one built utilizing an authoring tool (Software as a Service); and it demonstrates how a gastronomic business could attract with ease new customers by interacting with them using chatbots “embedded” into instant messaging apps.
Download
Paper Nr: 78
Title:

Automated Attribute Inference for IoT Data Visualization Service
Authors:

Orathai Sangpetch, Akkarit Sangpetch, Jittinat Nartnorakij and Narawan Vejprasitthikul
Abstract: As data becomes vital to urban development of modern cities, Thailand has initiated a smart city project on pilot cities around the country. We have implemented an interoperable data platform for smart city to enable Internet of Things (IoT) data exchanges among organizations through APIs. One of the key success is that people can access and visual the data. However, data can have various attributes since standard has not completely established and adopted. Therefore, it is difficult to automate the process to achieve comprehensive visualization. Traditionally, we require developers to manually examine data streams to determine which data attribute should be presented. This process can be very time consuming. The visualization system must be manually updated whenever a source stream modifies its data attributes. This problem becomes an impediment to implement a scalable cloud-based visualization service. To mitigate this challenge, we propose an automated attribute inference approach to automatically select key visualizable attribute from heterogeneous streams of data sources. We have experimented with different data attribute selection algorithms, namely an empirical rule-based system and the chosen machine learning algorithms. We implement and evaluate the proposed selection algorithms through our 3D visualization program in order to get the feedback from users.
Download

Area 9 - Cloud Computing Enabling Technology

Full Papers
Paper Nr: 12
Title:

On Incorporating Security Parameters in Service Level Agreements
Authors:

Aida Čaušević, Elena Lisova, Mohammad Ashjaei and Syed U. Ashgar
Abstract: With development of cloud computing new ways for easy, on-demand, Internet-based access to computing resources have emerged. In such context a Service Level Agreement (SLA) enables contractual agreements between service providers and users. Given an SLA, service users are able to establish trust in that the service outcome corresponds to what they have demanded during the service negotiation process. However, an SLA provides a limited support outside of basic Quality of Service (QoS) parameters, especially when it comes to security. We find security as an important factor to be included in adjusting an SLA according to user defined objectives. Incorporating it in an SLA is challenging due to difficulty to provide complete and quantifiable metrics, thus we propose to focus on a systematic way of addressing security using the security process. In this paper we investigate ways in which security might be incorporated already in the service negotiation process and captured in an SLA. We propose a corresponding process to develop and maintain an SLA that considers both design-, and run-time. To demonstrate the approach we built upon the existing SLAC language and extend its syntax to support security. An example of a service being provided with security guarantees illustrates the concept.
Download
Paper Nr: 14
Title:

Efficient Task Scheduling in Cloud Computing using an Improved Particle Swarm Optimization Algorithm
Authors:

Guang Peng and Katinka Wolter
Abstract: An improved multi-objective discrete particle swarm optimization (IMODPSO) algorithm is proposed to solve the task scheduling and resource allocation problem for scientific workflows in cloud computing. First, we use a strategy to limit the velocity of particles and adopt a discrete position updating equation to solve the multi-objective time and cost optimization model. Second, we adopt a Gaussian mutation operation to update the personal best position and the external archive, which can retain the diversity and convergence accuracy of Pareto optimal solutions. Finally, the computational complexity of IMODPSO is compared with three other state-of-the-art algorithms. We validate the computational speed, the number of solutions found and the generational distance of IMODPSO and find that the new algorithm outperforms the three other algorithms with respect to all three metrics.
Download
Short Papers
Paper Nr: 18
Title:

Framework Node2FaaS: Automatic NodeJS Application Converter for Function as a Service
Authors:

Leonardo Rebouças de Carvalho and Aletéia P. Favacho de Araújo
Abstract: Cloud computing emerged in the area of computer science as a means to achieve significant cost and time savings when starting projects. Among the various cloud models available, this work highlights Function as a Service - FaaS, and proposes the Node2FaaS framework for automatic conversion of applications written in NodeJS to work in a transparent way with the FaaS model. The experiments demonstrated significant gains of up to 170% at runtime for applications with high file I/O requirements. Applications with high CPU and RAM consumption also have benefits in adopting FaaS after conversion, but only when a threshold of competing processes is reached.
Download
Paper Nr: 30
Title:

A Deep-learning-based approach to VM behavior Identification in Cloud Systems
Authors:

Matteo Stefanini, Riccardo Lancellotti, Lorenzo Baraldi and Simone Calderara
Abstract: Cloud computing data centers are growing in size and complexity to the point where monitoring and management of the infrastructure become a challenge due to scalability issues. A possible approach to cope with the size of such data centers is to identify VMs exhibiting a similar behavior. Existing literature demonstrated that clustering together VMs that show a similar behavior may improve the scalability of both monitoring and management of a data center. However, available clustering techniques suffer from a trade-off between the accuracy of the clustering and the time to achieve this result. Not being able to obtain an accurate clustering in short time hinders the application of these solutions, especially in public cloud scenarios where on-demand VMs are instantiated and run for a short time span. Throughout this paper we propose a different approach where, instead of an unsupervised clustering, we rely on classifiers based on deep learning techniques to assign a newly deployed VMs to a cluster of already-known VMs. The two proposed classifiers, namely DeepConv and DeepFFT use a convolution neural network and (in the latter model) exploits Fast Fourier Transformation to classify the VMs. Our proposal is validated using a set of traces describing the behavior of VMs from a real cloud data center. The experiments compare our proposal with state-of-the-art solutions available in literature, such as the AGATE technique and PCA-based clustering, demonstrating that our proposal can achieve a very high accuracy (compared to the best performing alternatives) without the need to introduce the notion of a gray-area to take into account not-yet assigned VMs as in AGATE. Furthermore, we show that our solution is significantly faster than the alternatives as it can produce a perfect classification even with just a few samples of data, such as 4 observations (corresponding to 20 minutes of data), making our proposal viable also to classify on-demand VMs that are characterized by a short life span.
Download
Paper Nr: 54
Title:

Towards Balancing Energy Savings and Performance for Volunteer Computing through Virtualized Approach
Authors:

Fábio Rossi, Tiago Ferreto, Marcelo Conterato, Paulo Souza, Wagner Marques, Rodrigo Calheiros and Guilherme Rodrigues
Abstract: Computational grids consist of distributed environments where partner institutions offer hosts along with computational resources that can be used by all members of the grid. When an application needs to run in such environment, it allocates a portion of hosts necessary for its executions. Traditionally, the workload imposed on computational grids has a characteristic of being bag-of-tasks (BoT). It means that multiple replicas are submitted to different hosts, and when a response is processed, such replicas are either ignored or released. On resource allocation, as the grid is distributed among different participants, only idle resources can be leased by a new application. However, due to the behavior of BoTs, many allocated resources do not use their resources in their entirety. Another important fact is that only fully idle hosts can be added to the grid pool, and used only at these times. From the above, this paper proposes an approach that uses underutilized resource slice of grid hosts through virtualization, adjusting the use of grid applications to the leftover resources from daily hosts usage. It allows grid applications to run, even when hosts are in use, as long as there is an idle slice of resources, and their use does not interfere with the host’s current running. To evaluate this approach, we performed an empirical evaluation of a virtualized server running applications concurrently with a virtualized grid application. The results have showed that our scheme could accelerate the performance of grid applications without impacting on higher energy consumption.
Download
Paper Nr: 61
Title:

Machine Learning Approach for Live Migration Cost Prediction in VMware Environments
Authors:

Mohamed E. Elsaid, Hazem M. Abbas and Christoph Meinel
Abstract: Virtualization became a commonly used technology in datacenters during the last decade. Live migration is an essential feature in most of the clusters hypervisors. Live migration process has a cost that includes the migration time, downtime, IP network overhead, CPU overhead and power consumption. This migration cost cannot be ignored, however datacenter admins do live migration without expectations about the resultant cost. Several research papers have discussed this problem, however they could not provide a practical model that can be easily implemented for cost prediction in VMware environments. In this paper, we propose a machine learning approach for live migration cost prediction in VMware environments. The proposed approach is implemented as a VMware PowerCLI script that can be easily implemented and run in any vCenter Server Cluster to do data collection of previous migrations statistics, train the machine learning models and then predict live migration cost. Testing results show how the proposed framework can predict live migration time, network throughput and power consumption cost with accurate results and for different kinds of workloads. This helps datacenters admins to have better planning for their VMware environments live migrations.
Download
Paper Nr: 62
Title:

Workspace-based Virtual Networks: A Clean Slate Approach to Slicing Cloud Networks
Authors:

Romerson D. Oliveira, Diego N. Molinos, Marcelo S. Freitas, Pedro F. Rosa and Flavio O. Silva
Abstract: Cloud Computing has brought a new vision about what applications might expect from underlying networks. Software Defined Networking, in this context, provides techniques to make networks more flexible to application requirements as well as to virtual networks, which help implement the provisioning of physical resources to support tenants. Cloud-hosted applications assume that networks have high throughput, but do not consider competition through the medium, handled by the switches. This paper aims to present the concept of Workspace as a new paradigm to manage the underlying resources in terms of meeting the communication requirements. We introduce a clean-slate approach based on horizontal addressing that has Workspaces as logical links capable of being parameterized at the level of medium access control. In addition, an overview of a new network element designed to allow the parameterization of Workspaces directly on the hardware is given.
Download
Paper Nr: 75
Title:

Proof of Provision: Improving Blockchain Technology by Cloud Computing
Authors:

Matthias Pohl, Abdulrahman Nahhas, Sascha Bosse and Klaus Turowski
Abstract: Blockchain technology is mainly used in so-called cryptocurrencies and smart contracts. From a technological point of view, securing these applications requires a lot of computing power, which results in high energy consumption. The reason for this is the proof of work algorithm integrated in most cases. In order to promote the problem of high energy consumption and the sustainable use of cloud computing resources, this paper presents a consensus concept for use in a blockchain. We also present the classifications for discussion and give an outlook on a future evaluation.
Download
Paper Nr: 79
Title:

Rule-based Security Monitoring of Containerized Workloads
Authors:

Holger Gantikow, Christoph Reich, Martin Knahl and Nathan Clarke
Abstract: In order to further support the secure operation of containerized environments and to extend already established security measures, we propose a rule-based security monitoring, which can be used for the detection of a variety of misuse and attacks. The capabilities of the open-source tools used to monitor containers are closely examined and the possibility of detecting undesired behavior is evaluated on the basis of various scenarios. Further, the limits of the approach taken and the associated performance overhead will be discussed. The results show that the proposed approach is effective in many scenarios and comes at a low performance overhead cost.
Download
Paper Nr: 93
Title:

IoT-DMCP: An IoT Data Management and Control Platform for Smart Cities
Authors:

Sahar Boulkaboul, Djamel Djenouri, Sadmi Bouhafs and Mohand N. Belaid
Abstract: This paper presents a design and implementation of a data management platform to monitor and control smart objects in the Internet of Things (IoT). This is through IPv4/IPv6, and by combining IoT specific features and protocols such as CoAP, HTTP and WebSocket. The platform allows anomaly detection in IoT devices and real-time error reporting mechanisms. Moreover, the platform is designed as a standalone application, which targets at extending cloud connectivity to the edge of the network with fog computing. It extensively uses the features and entities provided by the Capillary Networks with a micro-services based architecture linked via a large set of REST APIs, which allows developing applications independently of the heterogeneous devices. The platform addresses the challenges in terms of connectivity, reliability, security and mobility of the Internet of Things through IPv6. The implementation of the platform is evaluated in a smart home scenario and tested via numeric results. The results show low latency, at the order of few ten of milliseconds, for building control over the implemented mobile application, which confirm realtime feature of the proposed solution.
Download
Paper Nr: 37
Title:

Performance Analysis of an Hyperconverged Infrastructure using Docker Containers and GlusterFS
Authors:

Rodrigo Leite, Priscila Solis and Eduardo Alchieri
Abstract: The adoption of hyperconverged infrastructures is a trend in datacenters, because it merges different type of computing and storage resources. Hyperconverged infrastructures use distributed file systems (DFS) to store and replicate data between multiple servers while using computing resources of the same servers to host virtual machines or containers. In this work, the distributed file system GlusterFS and the hypervisor VMware ESXi are used to build an hyperconverged system to host Docker containers, with the goal of evaluate the storage performance of this system compared to traditional approach where data is stored directly on the server’s disks. The performance of the container’s persistent storage is evaluated using the benchmark tool Yahoo Cloud Service Benchmark (YCSB) against the NoSQL databases Cassandra and MongoDB under differents workloads. The NoSQL database’s performance was compared between the hyperconverged system with multiples disk configurations and a traditional system with local storage.
Download
Paper Nr: 63
Title:

Toward an Autonomic and Adaptive Load Management Strategy for Reducing Energy Consumption under Performance Constraints in Data Centers
Authors:

Abdulrahman Nahhas, Sascha Bosse, Matthias Pohl and Klaus Turowski
Abstract: The future vision of IT-industry is shifting toward a utility-based offering of computing power using the concepts of pay-per-use. However, the elasticity and scalability characteristics of cloud computing massively increased the complexity of IT-system landscapes, since market leaders extensively expanding their IT-infrastructure. Accordingly, the carbon-footprint of data centers operations is estimated to be the fastest growing footprint among different IT fields. The majority of contribution in the examined literature that address IT resources management in data centers exhibits either a specific or a generic nature. The specific solutions are designed to solve specific problems, but yet neglecting the dynamic nature of IT-systems. The design of generic solutions usually overlooks many details of the investigated problems that have an impact on the possible optimization potential. One can argue that an optimized combination of different algorithms used during a specified time span would outperform a single specific or generic algorithm for the management of IT recourses in data centers. Therefore, a conceptual design for an autonomic and adaptive load management strategy is presented to investigate the aforementioned hypothesis. Our initial experimental results showed considerable improvement when multiple algorithms are used for the allocation of virtual machines.
Download