Stelios Sotiriadis

An inter-cloud bridge system for heterogeneous cloud platforms

Over the years, more cloud computing systems have been developed providing flexible interfaces for inter-cloud interaction. This work approaches the concept of inter-cloud by utilizing APIs, open source specifications and exposed interfaces from cloud platforms such as OpenStack, OpenNebula and others. Despite other works in the area of inter-cloud, that are mainly resource management-centric, we focus on designing and developing a service-centric architecture. We implement an inter-cloud bridge system that is elastic, easy to be upgraded and managed. We develop a prototype composed not only from heterogeneous cloud platforms but also from independent cloud services. These are developed by different cloud service providers and offered as open source Software as a Service (SaaS). The proposed Inter-Cloud Mediation Service uses Future Internet SaaS such as a Context Broker for registrations and subscriptions to services and a Complex Event Processing engine for event management. We present an experimental analysis to show interactions with various heterogeneous cloud platforms and we evaluate the performance of inter-cloud services separately and as a whole. We model a service-centric service for Inter-Cloud communication.The architecture relies on different cloud providers and their open SaaS.We examine Inter-Cloud using cloud platform API interfaces (OpenStack and FIWARE).Approach shows efficiency regarding service stability and minimizes delays.Experiments show effective Inter-Cloud request, response and deployment times.

An Inter-Cloud Meta-Scheduling (ICMS) Simulation Framework: Architecture and Evaluation

Inter-cloud is an approach that facilitates scalable resource provisioning across multiple cloud infrastructures. In this paper, we focus on the performance optimization of Infrastructure as a Service (IaaS) using the meta-scheduling paradigm to achieve an improved job scheduling across multiple clouds. We propose a novel inter-cloud job scheduling framework and implement policies to optimize performance of participating clouds. The framework, named as Inter-Cloud Meta-Scheduling (ICMS), is based on a novel message exchange mechanism to allow optimization of job scheduling metrics. The resulting system offers improved flexibility, robustness and decentralization. We implemented a toolkit named “Simulating the Inter-Cloud” (SimIC) to perform the design and implementation of different inter-cloud entities and policies in the ICMS framework. An experimental analysis is produced for job executions in inter-cloud and a performance is presented for a number of parameters such as job execution, makespan, and turnaround times. The results highlight that the overall performance of individual clouds for selected parameters and configuration is improved when these are brought together under the proposed ICMS framework.

A Survey on Approaches for Interoperability and Portability of Cloud Computing Services

Over the recent years, the rapid development of Cloud Computing has driven to a large market of cloud n nservices that offer infrastructure, platforms and software to everyday users. Yet, due to the lack of common n naccepted standards, cloud service providers use different technologies and offer their clients services that are n noperated by a variety of proprietary APIs. The lack of standardization results in numerous heterogeneities n n(e.g., heterogeneous service descriptions, message level naming conflicts, data representation conflicts etc.) n nmaking the interoperation, collaboration and portability of services a very complex task. In this work we n nfocus on the problems of interoperability and portability in Cloud Computing, we address their differences n nand we discuss some of the latest research work in this area. We evaluate and point out relationships between n nthe identified solutions. Finally we present a use case scenario of the FI-STAR project, that aims to bridge the n ngap of healthcare provision in cloud computing. We demonstrate the architecture of the project and we discuss n npossible interoperability and portability issues.

Analysis of power consumption in heterogeneous virtual machine environments

Reduction of energy consumption in Cloud computing datacenters today is a hot a research topic, as these consume large amounts of energy. Furthermore, most of the energy is used inefficiently because of the improper usage of computational resources such as CPU, storage and network. A good balance between the computing resources and performed workload is mandatory. In the context of data-intensive applications, a significant portion of energy is consumed just to keep alive virtual machines or to move data around without performing useful computation. Moreover, heterogeneity of resources increases the difficulty degree, when trying to achieve energy efficiency. Power consumption optimization requires identification of those inefficiencies in the underlying system and applications. Based on the relation between server load and energy consumption, we study the efficiency of data-intensive applications, and the penalties, in terms of power consumption, that are introduced by different degrees of heterogeneity of the virtual machines characteristics in a cluster.

Self managed virtual machine scheduling in Cloud systems

In Cloud systems, Virtual Machines (VMs) are scheduled to hosts according to their instant resource usage (e.g. to hosts with most available RAM) without considering their overall and long-term utilization. Also, in many cases, the scheduling and placement processes are computational expensive and affect performance of deployed VMs. In this work, we present a Cloud VM scheduling algorithm that takes into account already running VM resource usage over time by analyzing past VM utilization levels in order to schedule VMs by optimizing performance. We observe that Cloud management processes, like VM placement, affect already deployed systems (for example this could involve throughput drop in a database cluster), so we aim to minimize such performance degradation. Moreover, overloaded VMs tend to steal resources (e.g. CPU) from neighbouring VMs, so our work maximizes VMs real CPU utilization. Based on these, we provide an experimental analysis to compare our solution with traditional schedulers used in OpenStack by exploring the behaviour of different NoSQL (MongoDB, Apache Cassandra and Elasticsearch). The results show that our solution refines traditional instant-based physical machine selection as it learns the system behaviour as well as it adapts over time. The analysis is prosperous as for the selected setting we approximately minimize performance degradation by 19% and we maximize CPU real time by 2% when using real world workloads.

An Inter-Cloud Architecture for Future Internet Infrastructures

In latest years, the concept of interconnecting clouds to allow common service coordination has gained significant attention mainly because of the increasing utilization of cloud resources from Internet users. An efficient common management between different clouds is essential benefit, like boundless elasticity and scalability. Yet, issues related with different standards led to interoperability problems. For this reason, the definition of the open cloud-computing interface defines a set of open community-lead specifications along with a flexible API to build cloud systems. Today, there are cloud systems like OpenStack, OpenNebula, Amazon Web Services and VMWare VCloud that expose APIs for inter-cloud communication. In this work we aim to explore an inter-cloud model by creating a new cloud platform service to act as a mediator among OpenStack, FI-WARE datacenter resource management and Amazon Web Service cloud architectures, therefore to orchestrate communication of various cloud environments. The model is based on the FI-WARE and will be offered as a reusable enabler with an open specification to allow interoperable service coordination.

Modular and generic IoT management on the cloud

Cloud computing and Internet of Things encompass various physical devices that generate and exchange data with services promoting the integration between the physical world and computer-based systems. This work presents a novel Future Internet cloud service for data collection from Internet of Things devices in an automatic, generalized and modular way. It includes a flexible API for managing devices, users and permissions by mapping data to users, publish and subscribe context data as well as storage capabilities and data processing in the form of NoSQL big data. The contributions of this work include the on the fly data collection from devices that is stored in cloud scalable databases, the vendor agnostic Internet of Things device connectivity (since it is designed to be flexible and to support device heterogeneity), and finally the modularity of the event based publish/subscribe service for context oriented data that could be easily utilized by third party services without worrying about how data are collected, stored and managed.

IoT-A and FIWARE: Bridging the Barriers between the Cloud and IoT Systems Design and Implementation

Today, IoT systems are designed and implemented to address specific challenges based on domain specific requirements, thus not taking into consideration issues of openness, scalability, interoperability and use-case independence. As a result, they are less principled, lacking standards, vendor oriented and hardly replicable since the same IoT architecture cannot be used in more than one use-cases. To address the fragmentation of existing IoT solutions, the IoT-A project proposes an architecture reference model that defines the principles and standards for generating IoT architectures and promoting the interoperation of IoT solutions. However, IoT-A addresses the architecture design problem, and does not focus on whether existing cloud platforms can offer the tools and services to support the implementation of IoT-A compliant IoT systems. In this work we propose an architecture based on IoT-A that focuses on the FIWARE open cloud platform that in turn provides the building blocks of future Internet applications and services. We further correlate FIWARE and IoT-A projects to identify the key features for FIWARE to support IoT-A compliant system implementations.

Semantic Aware Online Detection of Resource Anomalies on the Cloud

As cloud based platforms become more popular, it becomes an essential task for the cloud administrator to efficiently manage the costly hardware resources in the cloud environment. Prompt action should be taken whenever hardware resources are faulty, or configured and utilized in a way that causes application performance degradation, hence poor quality of service. In this paper, we propose a semantic aware technique based on neural network learning and pattern recognition in order to provide automated, real-time support for resource anomaly detection. We incorporate application semantics to narrow down the scope of the learning and detection phase, thus enabling our machine learning technique to work at a very low overhead when executed online. As our method runs life-long on monitored resource usage on the cloud, in case of wrong prediction, we can leverage administrator feedback to improve prediction on future runs. This feedback directed scheme with the attached context helps us to achieve an anomaly detection accuracy of as high as 98.3% in our experimental evaluation, and can be easily used in conjunction with other anomaly detection techniques for the cloud.

Vertical and horizontal elasticity for dynamic virtual machine reconfiguration

Today, cloud computing applications are rapidly constructed by services belonging to different cloud providers and service owners. This work presents the inter-cloud elasticity framework, which focuses on cloud load balancing based on dynamic virtual machine reconfiguration when variations on load or on user requests volume are observed. We design a dynamic reconfiguration system, called inter-cloud load balancer (ICLB), that allows scaling up or down the virtual resources (thus providing automatized elasticity), by eliminating service downtimes and communication failures. It includes an inter-cloud load balancer for distributing incoming user HTTP traffic across multiple instances of inter-cloud applications and services and we perform dynamic reconfiguration of resources according to the real time requirements. The experimental analysis includes different topologies by showing how real-time traffic variation (using real world workloads) affects resource utilization and by achieving better resource usage in inter-cloud.