Simulation of SLA-based VM-scaling algorithms for cloud-distributed applications

Cloud Computing has evolved to become an enabler for delivering access to large scale distributed applications running on managed network-connected computing systems. This makes possible hosting Distributed Enterprise Information Systems (dEISs) in cloud environments, while enforcing strict performance and quality of service requirements, defined using Service Level Agreements (SLAs). {SLAs} define the performance boundaries of distributed applications, and are enforced by a cloud management system (CMS) dynamically allocating the available computing resources to the cloud services. We present two novel VM-scaling algorithms focused on dEIS systems, which optimally detect most appropriate scaling conditions using performance-models of distributed applications derived from constant-workload benchmarks, together with SLA-specified performance constraints. We simulate the VM-scaling algorithms in a cloud simulator and compare against trace-based performance models of dEISs. We compare a total of three SLA-based VM-scaling algorithms (one using prediction mechanisms) based on a real-world application scenario involving a large variable number of users. Our results show that it is beneficial to use autoregressive predictive SLA-driven scaling algorithms in cloud management systems for guaranteeing performance invariants of distributed cloud applications, as opposed to using only reactive SLA-based VM-scaling algorithms.

Mobility and Bandwidth Prediction as a Service in Virtualized LTE Systems

Recently telecommunication industry benefits from infrastructure sharing, one of the most fundamental enablers of cloud computing, leading to emergence of the Mobile Virtual Network Operator (MVNO) concept. The most momentous intents by this approach are the support of on-demand provisioning and elasticity of virtualized mobile network components, based on data traffic load. To realize it, during operation and management procedures, the virtualized services need be triggered in order to scale-up/down or scale-out/in an instance. In this paper we propose an architecture called MOBaaS (Mobility and Bandwidth Availability Prediction as a Service), comprising two algorithms in order to predict user(s) mobility and network link bandwidth availability, that can be implemented in cloud based mobile network structure and can be used as a support service by any other virtualized mobile network services. MOBaaS can provide prediction information in order to generate required triggers for on-demand deploying, provisioning, disposing of virtualized network components. This information can be used for self-adaptation procedures and optimal network function configuration during run-time operation, as well. Through the preliminary experiments with the prototype implementation on the OpenStack platform, we evaluated and confirmed the feasibility and the effectiveness of the prediction algorithms and the proposed architecture.

Trends in Human-Centric Multimedia Networking Scenarios

The evolution of wireless access technologies and mobile devices, together with the constant demand for video services, has created new Human-Centric Multimedia Networking (HCMN) scenarios. However, HCMN poses several challenges for content creators and network providers to deliver multimedia data with an acceptable quality level based on the user experience. Moreover, human experience and context, as well as network information play an important role in adapting and optimizing video dissemination. In this paper, we discuss trends to provide video dissemination with Quality of Experience (QoE) support by integrating HCMN with cloud computing approaches. We identified five trends coming from such integration, namely Participatory Sensor Networks, Mobile Cloud Computing formation, QoE assessment, QoE management, and video or network adaptation.

Testing a Cloud Provider Network for Hybrid P2P and Cloud Streaming Architectures

The number of online real-time streaming services deployed over network topologies like P2P or centralized ones has remarkably increased in the recent years. This has revealed the lack of networks that are well prepared to respond to this kind of traffic. A hybrid distribution network can be an efficient solution for real-time streaming services. This paper contains the experimental results of streaming distribution in a hybrid architecture that consist of mixed connections among P2P and Cloud nodes that can interoperate together. We have chosen to represent the P2P nodes as Planet Lab machines over the world and the cloud nodes using a Cloud provider's network. First we present an experimental validation of the Cloud infrastructure's ability to distribute streaming sessions with respect to some key streaming QoS parameters: jitter, throughput and packet losses. Next we show the results obtained from different test scenarios, when a hybrid distribution network is used. The scenarios measure the improvement of the multimedia QoS parameters, when nodes in the streaming distribution network (located in different continents) are gradually moved into the Cloud provider infrastructure. The overall conclusion is that the QoS of a streaming service can be efficiently improved, unlike in traditional P2P systems and CDN, by deploying a hybrid streaming architecture. This enhancement can be obtained by strategic placing of certain distribution network nodes into the Cloud provider infrastructure, taking advantage of the reduced packet loss and low latency that exists among its datacenters.

CumuloNimbo: Una plataforma como servicio con procesamiento transaccional altamente escalable.

El modelo de computaci¿on en la nube (cloud computing) ha ganado mucha popularidad en los últimos años, prueba de ello es la cantidad de productos que distintas empresas han lanzado para ofrecer software, capacidad de procesamiento y servicios en la nube. Para una empresa el mover sus aplicaciones a la nube, con el fin de garantizar disponibilidad y escalabilidad de las mismas y un ahorro de costes, no es una tarea fácil. El principal problema es que las aplicaciones tienen que ser rediseñadas porque las plataformas de computaci¿on en la nube presentan restricciones que no tienen los entornos tradicionales. En este artículo presentamos CumuloNimbo, una plataforma para computación en la nube que permite la ejecución y migración de manera transparente de aplicaciones multi-capa en la nube. Una de las principales características de CumuloNimbo es la gestión de transacciones altamente escalable y coherente. El artículo describe la arquitectura del sistema, así como una evaluaci¿on de la escalabilidad del mismo.

A federated repository for PaaS components in a multi-cloud environment

Cloud computing has seen an impressive growth in recent years, with virtualization technologies being massively adopted to create IaaS (Infrastructure as a Service) public and private solutions. Today, the interest is shifting towards the PaaS (Platform as a Service) model, which allows developers to abstract from the execution platform and focus only on the functionality. There are several public PaaS offerings available, but currently no private PaaS solution is ready for production environments. To fill this gap a new solution must be developed. In this paper we present a key element for enabling this model: a cloud repository based on the OSGi component model. The repository stores, manages, provisions and resolves the dependencies of PaaS software components and services. This repository can federate with other repositories located in the same or different clouds, both private and public. This way, dependencies can be fulfilled collaboratively, and new business models can be implemented.

Tecnología cloud para el hogar digital

El mundo tecnológico está cambiando hacia la optimización en la gestión de recursos gracias a la poderosa influencia de tecnologías como la virtualización y la computación en la nube (Cloud Computing). En esta memoria se realiza un acercamiento a las mismas, desde las causas que las motivaron hasta sus últimas tendencias, pasando por la identificación de sus principales características, ventajas e inconvenientes. Por otro lado, el Hogar Digital es ya una realidad para la mayoría de los seres humanos. En él se dispone de acceso a múltiples tipos de redes de telecomunicaciones (3G, 4G, WI-FI, ADSL…) con más o menos capacidad pero que permiten conexiones a internet desde cualquier parte, en todo momento, y con prácticamente cualquier dispositivo (ordenadores personales, smartphones, tabletas, televisores…). Esto es aprovechado por las empresas para ofrecer todo tipo de servicios. Algunos de estos servicios están basados en el cloud computing sobre todo ofreciendo almacenamiento en la nube a aquellos dispositivos con capacidad reducida, como son los smarthphones y las tabletas. Ese espacio de almacenamiento normalmente está en los servidores bajo el control de grandes compañías. Guardar documentos, videos, fotos privadas sin tener la certeza de que estos no son consultados por alguien sin consentimiento, puede despertar en el usuario cierto recelo. Para estos usuarios que desean control sobre su intimidad, se ofrece la posibilidad de que sea el propio usuario el que monte sus propios servidores y su propio servicio cloud para compartir su información privada sólo con sus familiares y amigos o con cualquiera al que le dé permiso. Durante el proyecto se han comparado diversas soluciones, la mayoría de código abierto y de libre distribución, que permiten desplegar como mínimo un servicio de almacenamiento accesible a través de Internet. Algunas de ellas lo complementan con servicios de streaming tanto de música como de videos, compartición y sincronización de documentos entre múltiples dispositivos, calendarios, copias de respaldo (backups), virtualización de escritorios, versionado de ficheros, chats, etc. El proyecto finaliza con una demostración de cómo utilizar dispositivos de un hogar digital interactuando con un servidor Cloud, en el que previamente se ha instalado y configurado una de las soluciones comparadas. Este servidor quedará empaquetado en una máquina virtual para que sea fácilmente transportable e utilizable. ABSTRACT. The technological world is changing towards optimizing resource management thanks to the powerful influence of technologies such as Virtualization and Cloud Computing. This document presents a closer approach to them, from the causes that have motivated to their last trends, as well as showing their main features, advantages and disadvantages. In addition, the Digital Home is a reality for most humans. It provides access to multiple types of telecommunication networks (3G, 4G, WI-FI, ADSL...) with more or less capacity, allowing Internet connections from anywhere, at any time, and with virtually any device (computer personal smartphones, tablets, televisions...).This is used by companies to provide all kinds of services. Some of these services offer storage on the cloud to devices with limited capacity, such as smartphones and tablets. That is normally storage space on servers under the control of important companies. Saving private documents, videos, photos, without being sure that they are not viewed by anyone without consent, can wake up suspicions in some users. For those users who want control over their privacy, it offers the possibility that it is the user himself to mount his own server and its own cloud service to share private information only with family and friends or with anyone with consent. During the project I have compared different solutions, most open source and with GNU licenses, for deploying one storage facility accessible via the Internet. Some supplement include streaming services of music , videos or photos, sharing and syncing documents across multiple devices, calendars, backups, desktop virtualization, file versioning, chats... The project ends with a demonstration of how to use our digital home devices interacting with a cloud server where one of the solutions compared is installed and configured. This server will be packaged in a virtual machine to be easily transportable and usable.

Valuation: Developer support for by-references to by-value type conversion.

Modern object oriented languages like C# and JAVA enable developers to build complex application in less time. These languages are based on selecting heap allocated pass-by-reference objects for user defined data structures. This simplifies programming by automatically managing memory allocation and deallocation in conjunction with automated garbage collection. This simplification of programming comes at the cost of performance. Using pass-by-reference objects instead of lighter weight pass-by value structs can have memory impact in some cases. These costs can be critical when these application runs on limited resource environments such as mobile devices and cloud computing systems. We explore the problem by using the simple and uniform memory model to improve the performance. In this work we address this problem by providing an automated and sounds static conversion analysis which identifies if a by reference type can be safely converted to a by value type where the conversion may result in performance improvements. This works focus on C# programs. Our approach is based on a combination of syntactic and semantic checks to identify classes that are safe to convert. We evaluate the effectiveness of our work in identifying convertible types and impact of this transformation. The result shows that the transformation of reference type to value type can have substantial performance impact in practice. In our case studies we optimize the performance in Barnes-Hut program which shows total memory allocation decreased by 93% and execution time also reduced by 15%.

An architecture for a heterogeneous private IaaS management system

Cloud computing and, more particularly, private IaaS, is seen as a mature technology with a myriad solutions tochoose from. However, this disparity of solutions and products has instilled in potential adopters the fear of vendor and data lock-in. Several competing and incompatible interfaces and management styles have given even more voice to these fears. On top of this, cloud users might want to work with several solutions at the same time, an integration that is difficult to achieve in practice. In this paper, we propose a management architecture that tries to tackle these problems; it offers a common way of managing several cloud solutions, and an interface that can be tailored to the needs of the user. This management architecture is designed in a modular way, and using a generic information model. We have validated our approach through the implementation of the components needed for this architecture to support a sample private IaaS solution: OpenStack

An online failure prediction system for private IaaS platforms

The size and complexity of cloud environments make them prone to failures. The traditional approach to achieve a high dependability for these systems relies on constant monitoring. However, this method is purely reactive. A more proactive approach is provided by online failure prediction (OFP) techniques. In this paper, we describe a OFP system for private IaaS platforms, currently under development, that combines di_erent types of data input, including monitoring information, event logs, and failure data. In addition, this system operates at both the physical and virtual planes of the cloud, taking into account the relationships between nodes and failure propagation mechanisms that are unique to cloud environments.

Transactional failure recovery for a distributed key-value store

With the advent of cloud computing, many applications have embraced the ensuing paradigm shift towards modern distributed key-value data stores, like HBase, in order to benefit from the elastic scalability on offer. However, many applications still hesitate to make the leap from the traditional relational database model simply because they cannot compromise on the standard transactional guarantees of atomicity, isolation, and durability. To get the best of both worlds, one option is to integrate an independent transaction management component with a distributed key-value store. In this paper, we discuss the implications of this approach for durability. In particular, if the transaction manager provides durability (e.g., through logging), then we can relax durability constraints in the key-value store. However, if a component fails (e.g., a client or a key-value server), then we need a coordinated recovery procedure to ensure that commits are persisted correctly. In our research, we integrate an independent transaction manager with HBase. Our main contribution is a failure recovery middleware for the integrated system, which tracks the progress of each commit as it is flushed down by the client and persisted within HBase, so that we can recover reliably from failures. During recovery, commits that were interrupted by the failure are replayed from the transaction management log. Importantly, the recovery process does not interrupt transaction processing on the available servers. Using a benchmark, we evaluate the impact of component failure, and subsequent recovery, on application performance.

Development of a management system for virtual machines on private clouds

Cloud computing and, more particularly, private IaaS, is seen as a mature technol- ogy with a myriad solutions to choose from. However, this disparity of solutions and products has instilled in potential adopters the fear of vendor and data lock- in. Several competing and incompatible interfaces and management styles have increased even more these fears. On top of this, cloud users might want to work with several solutions at the same time, an integration that is difficult to achieve in practice. In this Master Thesis I propose a management architecture that tries to solve these problems; it provides a generalized control mechanism for several cloud infrastructures, and an interface that can meet the requirements of the users. This management architecture is designed in a modular way, and using a generic infor- mation model. I have validated the approach through the implementation of the components needed for this architecture to support a sample private IaaS solution: OpenStack.

CloudRoom: módulo de gestión de cursos online, abiertos y masivos

La razón de este proyecto, es la de desarrollar el módulo de cursos de la plataforma de Massive Online Open Courses (MOOCs), CloudRoom. Dicho módulo está englobado en una arquitectura orientada a servicios (SOA) y en una infraestructura de Cloud Computing utilizando Amazon Web Services (AWS). Nuestro objetivo es el de diseñar un Software as a Service (SaaS) robusto con las cualidades que a un producto de este tipo se le estiman: alta disponibilidad, alto rendimiento, gran experiencia de usuario y gran extensibilidad del sistema. Para lograrlo, se llevará a cabo la integración de las últimas tendencias tecnológicas dentro del desarrollo de sistemas distribuidos como Neo4j, Node.JS, Servicios RESTful, CoffeeScript. Todo esto siguiendo un estrategia de desarrollo PLAN-DO-CHECK utilizando Scrum y prácticas de metodologías ágiles. ---ABSTRACT---The reason of this Project is to develop the courses‟ module of CloudRoom, a Massive Online Open Courses platform. This module is encapsulated in a service-oriented architecture (SOA) based on a Cloud Computing infrastructure built on Amazon Web Services (AWS). Our goal is to design a robust Software as a Service (SaaS) with the qualities that are estimated in a product of this type: high availability, high performance, great user experience and great extensibility of the system. In order to address this, we carry out the integration of the latest technology trends in the development of distributed systems: Neo4j, Node.JS, RESTful Services and CoffeeScript. All of this, following a development strategy PLAN-DO-CHECK, using Scrum and practices of agile methodologies.

Implementación de una herramienta de productividad para el desarrollo y despliegue de aplicaciones

En la actualidad se está viviendo el auge del Cloud Computing (Computación en la Nube) y cada vez son más las empresas importantes en el sector de las Tecnologías de la Información que apuestan con fuerza por estos servicios. Por un lado, algunas ofrecen servicios, como Amazon y su sistema IaaS (Infrastructure as a Service) Amazon Web Services (AWS); por otro, algunas los utilizan, como ocurre en el caso de este proyecto, en el que Telefonica I+D hace uso de los servicios proporcionados por AWS para sus proyectos. Debido a este crecimiento en el uso de las aplicaciones distribuidas es importante tener en cuenta el papel que desempeñan los desarrolladores y administradores de sistemas que han de trabajar y mantener todas las máquinas remotas de uno o varios proyectos desde una única máquina local. El ayudar a realizar estas tareas de la forma más cómoda y automática posible es el objetivo principal de este proyecto. En concreto, el objetivo de este proyecto es el diseño y la implementación de una solución software que ayude a la productividad en el desarrollo y despliegue de aplicaciones en un conjunto de máquinas remotas desde una única máquina local, teniendo como base una prueba de concepto realizada anteriormente que prueba las funcionalidades más básicas de las librerías utilizadas para el desarrollo de la herramienta. A lo largo de este proyecto se han estudiado las diferentes alternativas que se encuentran en el mercado que ofrecen al menos parte de la soluci6n a los problemas abordados, pese a que los requisitos de la empresa indicaban que la herramienta debía implementarse de forma completa. Se estudió a fondo después la prueba de concepto de la que se partía para, con los conocimientos adquiridos sobre el tema, mejorarla cumpliendo los objetivos marcados. Tras el desarrollo y la implementaci6n completa de la herramienta se proponen posibles caminos a seguir en el futuro. ---ABSTRACT---Nowadays we are experiencing the rise of Cloud Computing and every day more and more important IT companies are betting hard for this kind of services. On one hand, some of these companies offer services such as Amazon IaaS (Infrastructure as a Service) system Amazon Web Services (AWS); on the other hand, some of them use these services, as in the case of this project, in which Telefonica I+D uses the services provided by AWS in their projects. Due this growth in the use of distributed applications it is important to consider the developers and system administrators' roles, who have to work and do the maintenance of all the remote machines from one or several projects from a single local machine. The main goal of this project is to help with these tasks making them as comfortable and automatically as possible. Specifically, the goal of this project is the design and implementation of a software solution that helps to achieve a better productivity in the development of applications on a set of remote machines from a single local machine, based on a proof of concept developed before, in which the basic functionality of the libraries used in this tool were tested. Throughout this project the different alternatives on the market that offer at least part of the solution to the problem addressed have been studied, although according to the requirements of the company, the tool should be implemented from scratch. After that, the basic proof of concept was thoroughly studied and improved with the knowledge acquired on the subject, fulfilling the marked goals. Once the development and full implementation of the tool is done, some ways of improvement for the future are suggested.

Desarrollo de una solución SaaS para una Enterprise Store

En el marco del proyecto europeo FI-WARE, en el CoNWet Lab (laboratorio de la ETSI Informáticos de la UPM) se ha implementado la plataforma Web Wstore que es una implementación de referencia del Store Generic Enabler perteneciente a dicho proyecto. El objetivo de FI-WARE es crear la plataforma núcleo del Internet del Futuro (IoF) con la intención de incrementar la competitividad global europea en el mundo de las TI. El proyecto introduce una infraestructura innovadora para la creación y distribución de servicios digitales en internet. WStore ofrece a los proveedores de servicios la plataforma donde publicar sus ofertas y desde la cual los clientes podrán acceder ellas. Estos proveedores ofrecen servicios Web, aplicaciones, widgets y data sets del mismo modo que Google ofrece aplicaciones en la tienda online Google Play o Apple en el App Store. WStore está implementada actualmente como una plataforma Web, por lo que una organización que desee ofrecer el servicio de la store necesita instalar el software en un servidor propio y disponer de un dominio para ofrecer sus productos. El objetivo de este trabajo es migrar WStore a un entorno de computación en la nube de manera que con una única instancia se ofrezca el servicio a las organizaciones que deseen disponer de su propia plataforma, de la cual tendrán total control como si se encontrase en su propia infraestructura. Para esto se implementa una versión de WStore que será desplegada en una infraestructura cloud y ofrecida como Software as a Service. La implementación incluye una serie de módulos de código que se podrán añadir opcionalmente en el proceso de instalación si se desea que la instancia instalada sea Multitenant. Además, en este trabajo se estudian y prueban las herramientas que ofrece MongoDB para desplegar la plataforma Wstore Multitenant en una infraestructura cloud. Estas herramientas son replica sets y sharding que permiten desplegar una base de datos escalable y de alta disponibilidad. ---ABSTRACT---In the context of the European project FI-WARE, the CoNWeT Lab (IT Lab from ETSIINF UPM university) has been implemented the web platform WStore. WStore is a reference implementation of the Generic Enabler Store from FI-WARE project. The FI-WARE goal is to create the core platform of the Future Internet (IoF) with the intention of enhancing Europe's global competitiveness in IT technologies. FI-WARE introduces an innovative infrastructure for the creation and distribution of digital services over the Internet. WStore offers to service providers a platform to publicate offerings and where customers can access them. The providers offer web services, applications, widgets and data sets in the same way that Google offers online applications on Google Play or Apple on App Store plataforms. WStore is currently implemented as a web platform, so if an organization wants to offer the store service, it need to install the software on it’s own serves and have a domain to offer their products. The objective of this paper is to migrate WStore to a cloud computing environment where a single instance of the WStore is offered as a web service to organizations who want their own store. Customers (tenants) of the WStore web service will have total control over the software and WStore administration. The implementation includes several code modules that can be optionally added in the installation process to build a Multitenant instance. In addition, this paper review the tools that MongoDB provide for scalability and high availability (replica sets and sharding) with the purpose of deploying multi-tenant WStore on a cloud infrastructure.