El autor Pablo Fernandez ha publicado 20 artículo(s):

A Service Level Agreement (SLA) regulates the provisioning of a service by defining a set of guarantees. Each guarantee sets a Service Level Objective (SLO) on some service metrics, and optionally a compensation that is applied when the SLO is unfulfilled (the compensation would be a penalty) or overfulfilled (the compensation would be a reward). For instance, Amazon is penalised with a 10% in service credits if the availability of the Elastic Cloud Computing service drops below 99.95%.
Currently, there are software tools and research proposals that use the information about compensations to automate and optimise certain parts of the service management. However, they assume that compensations are well defined, which is too optimistic in some circumstances and can lead to undesirable situations. For example, an unbounded, automated penalty was discarded in 2005 by the UK Royal Mail company after causing a loss of 280 million pounds in one year and a half.

In the article «Automated Validation of Compensable SLAs», published in IEEE Transactions on Services Computing (Early Access), and available at https://doi.org/10.1109/TSC.2018.2885766, we aim at answering the question «How can compensations be automatically validated?». To this end, we build on the compensable SLA model proposed in a previous work to provide a technique that leverages constraint satisfaction problem solvers to automatically validate them. We also present a materialisation of the model in iAgree, a language to specify SLAs and a tooling support that implements our whole approach. Our proposal has been evaluated by modelling and analysing the compensations of 24 SLAs of real-world scenarios including 319 guarantee terms. As a result, our technique has proven to be useful for detecting mistakes that are typically derived not only from the manual specification of SLAs in natural language, but also from the complex nature of compensation definitions. Thus, we found that nine guarantees with compensations that were not properly defined in the original SLAs specified in natural language. Specifically, five were wrongly specified by Verizon, and four were wrongly specified by the outsourcing service hiring of the regional governments of: Northwest Territories of Canada, and Andalusia in Spain. Therefore, our proposal can pave the way for using compensable SLAs in a safer and more reliable way.

Autores: Carlos Müller / Antonio Manuel Gutierrez / Pablo Fernandez / Octavio Martín-Díaz / Manuel Resinas / Antonio Ruiz-Cortés / 
Palabras Clave: Analysis - Compensation - CSP - Penalty - Reward - SLA - validation - WS-Agreement

Autores: Rafael Fresno-Aranda / Pablo Fernández / Antonio Ruiz-Cortés / 
Palabras Clave: Analysis - API - automation - Capacity - microservices

Availability is a key property in computational services and, therefore, is guaranteed by Service Level Agreements (SLAs) from the majority infrastructure services, such as virtualization (Amazon EC2, Windows Azure, Google Cloud, Joyent, Rackspace, …) and storage (Amazon S3, Google Cloud Storage, …). These SLAs describe availability in natural language and there are important differences in the scope and penalties that each service provides. Furthermore, descriptions use specific domain terms so they are difficult to understand by service customers. These circumstances make that availability analysis is a tedious, error-prone and time-consuming task. In this paper, we describe in detail this problem and provide a first approach to deal with these SLAs supported on current SLA analysis techniques.

Autores: Antonio Manuel Gutíerrez-Fernández / Pablo Fernández / Manuel Resinas / Antonio Ruiz-Cortés / 
Palabras Clave: Availability - Cloud - IaaS - Service Level Agreements

Las crecientes capacidades de los dispositivos finales han llevado al despliegue masivo de aplicaciones móviles distribuidas. El éxito de estas aplicaciones depende en gran medida de la Calidad del Servicio (QoS) que ofrecen. Esta calidad es especialmente difícil de evaluar debido al gran número de entidades implicadas y a su heterogeneidad. Las herramientas actuales de evaluación no dan soporte a este tipo de aplicaciones, suelen centrarse en la evaluación de la QoS proporcionada por una sola entidad. Sin embargo, la QoS de las aplicaciones distribuidas no sólo depende de la QoS de cada entidad, también hay que evaluar las interacciones entre las entidades.En este artículo se presenta un framework, denominado Perses, para lanzar entornos virtuales que permitan simular y evaluar la ejecución de aplicaciones móviles distribuidas. Esta simulación proporciona resultados de la QoS alcanzada. Además, el framework se ha integrado en una metodología DevOps para automatizar su ejecución. Vídeo de presentación: https://youtu.be/wpIApe_sPFE.

Autores: Sergio Laso / Javier Berrocal / Pablo Fernández / Antonio Ruiz Cortés / Juan Manuel Murillo / 
Palabras Clave: Aplicaciones Móviles Distribuidas - Calidad de Servicio - DevOps - Entorno Virtual

A pesar de la importancia de los servicios en la economía, las tareas como la búsqueda, análisis de alternativas, y contratación de servicios en virtud de acuerdos de nivel de servicio (ANS), siguen realizándose manualmente. En la denominada Web de los servicios existen alternativas para facilitar la automatización de estas tareas basadas en diversos modelos conceptuales: genéricos como Linked USDL, o centrados en algún aspecto concreto, como
WS-Agreement con los ANS. Sin embargo, estos últimos contemplan principalmente sólo aspectos técnicos, sin proporcionar una semántica explícita a los términos del ANS ni cumplir los principios de la Web, dificultando su adopción y análisis automático.

En este artículo presentamos Linked USDL Agreement, una extensión de la familia de ontologías Linked USDL que proporciona facilidades para especificar, gestionar y compartir descripciones de ANS en la Web. Este modelo semántico evita los problemas de interoperabilidad y heterogeneidad de las especificaciones de ANS actuales. Además, dado que nuestro modelo sigue los principios de la Web de los datos, las descripciones de ANS generadas son fácilmente publicables, compartibles y analizables, sirviendo como soporte del ciclo de vida de los servicios.

Nuestra propuesta ha sido validada tanto sobre servicios Web tradicionales (e.g. computación en la nube), como sobre servicios no-computacionales (e.g. outsourcing de procesos de negocio). La comparación realizada con otras alternativas existentes, así como la implementación de una herramienta que facilita la creación, publicación, y análisis automático de documentos en Linked USDL Agreement, nos permite afirmar que nuestra propuesta es capaz de soportar
completamente la gestión del ciclo de vida de los ANS.

Autores: José María García / Pablo Fernandez / Carlos Pedrinaci / Manuel Resinas / Jorge S. Cardoso / Antonio Ruiz Cortés / 
Palabras Clave:

Quality of an open source software ecosystem (OSS ecosystem) is key for different ecosystem actors such as contributors or adopters. In fact, the consideration of several quality aspects(e.g., activeness, visibility, interrelatedness, etc.) as a whole may provide a measure of the healthiness of OSS ecosystems. The more health a OSS ecosystem is, the more and better contributors and adopters it will gather. Some research tools have been developed to gather specific quality information from open source community data sources. However, there exist no frameworks available that can be used to evaluate their quality as a whole in order to obtain the health of an OSS ecosystem. To assess the health of these ecosystems, we propose to adopt robust principles and methods from the Service Oriented Computing field.

Autores: Oscar Franco-Bedoya / Marc Oriol / Carlos Müller / Jordi Marco / Pablo Fernández / Manuel Resinas / Xavier Franch / Antonio Ruiz-Cortés / 
Palabras Clave:

Aiming to be as competitive as possible, organisations are always pursuing to improve their business processes applying corrective actions when needed. However, the actual analysis and decision making for those actions is typically a challenging task relying on extensive human-in-the-loop expertise. Specifically, this improvement process usually involves: (i) to analyse evidences to understand the current behavior; (ii) to decide the actual objectives (usually defined in Service Level Agreements -SLAs- based on intuition) and (iii) to establish the improvement plan. In this ongoing work, we aim to propose a data-driven and intuition-free methodology to define an SLA as a governance element that specifies the service level objectives in an explicit way. Such a methodology considers process performance indicators that are analysed by means of inference, optimization, and simulation techniques. In order to motivate and exemplify our work we address a Healthcare scenario.

Autores: Carlos Müller / Minsu Cho / Pablo Fernandez / Minseok Song / Manuel Resinas / Antonio Ruiz-Cortés / 
Palabras Clave:

Software architecture tendencies are shifting to a microser- vice paradigm. In this context, RESTful APIs are being established the standard of integration. API designer often identifies two key issues to be competitive in such growing market. On the one hand, the generation of accurate documentation of the behavior and capabilities of the API to promote its usage; on the other hand, the design of a pricing plan that fits into the potential API user’s needs. Besides the increasing number of API modeling alternatives is emerging, there is a lack of proposals on the definition of flexible pricing plans usually contained in the Service Level Agreements (SLAs). In this paper we propose two different modeling techniques for the de- scription of SLA in a RESTful API context: iAgree and SLA4OAI.

Autores: Antonio Gámez-Díaz / Pablo Fernandez / Antonio Ruiz-Cortés / 
Palabras Clave:

El constante ritmo de evolución de la computación en la nube hace necesario que los términos de los servicios relativos a su precio y facturación se puedan ajustar dinámicamente, considerando cambios en aspectos como las infraestructuras de los proveedores, requisitos de los clientes, o las políticas de descuentos, entre otros. En este contexto, los contratos o acuerdos con clientes (en inglés Customer Agreements o CA) se usan para regular la provisión de servicios en la nube, incluyendo en ellos información referida al nivel de servicio acordado en los Acuerdos de Nivel de Servicio (SLA en inglés), los modelos de precios o los términos de facturación. Aunque existen varias propuestas académicas y herramientas industriales que ofrecen soporte a la definición de modelos de precios y facturación para servicios en la nube, hay carencias importantes en el nivel de personalización e integración de herramientas de monitorización automática en este proceso de facturación.En este artículo presentamos una propuesta altamente flexible para definir el ciclo de vida y proceso de facturación, la cual se basa en CAs que incluyen no sólo términos relativos al SLA, sino también a los modelos de precios y facturación, los cuales son altamente personalizables y pueden contener también compensaciones (en forma de descuentos o cargos adicionales). Además, hemos desarrollado una herramienta de monitorización y análisis automático que ha sido validada en un escenario real en la industria. Finalmente, en el artículo analizamos el rendimiento de nuestra herramienta y comparamos nuestra propuesta con más de 50 herramientas industriales, destacando las ventajas obtenidas gracias al acercamiento basado en reglas que hemos empleado.

Autores: José María García / Octavio Martín-Díaz / Pablo Fernandez / Carlos Müller / Antonio Ruiz-Cortés / 
Palabras Clave: Billing - Cloud Services - customer agreements - Monitoring - Pricing - Service Level Agreements

Cloud service providers offer to their customers a variety of pricing policies, which range from the simple, yet widely used pay-as-you-go schema to complex discounted models. When executing the billing process, stakeholders have to consider usage metrics and service level objectives in order to obtain the correct billing and conform to the service level agreement in place. The more metrics, discount and compensations rules are added to the pricing schema, the more complex the billing generation results. In this paper we present a monitoring-based solution that enables the dynamically definition of both service level objectives and discount rules, so that providers can customise the billing generation process in terms of the service level agreement they offer. We validate our proposal in a real-world scenario, introducing a micro-service based software solution deployed in a Kubernetes cluster.

Autores: José María García / Octavio Martín-Díaz / Pablo Fernandez / Carlos Müller / Antonio Ruiz-Cortés / 
Palabras Clave: Billing - Cloud Services - Monitoring - Pricing - Service Level Agreements

Autores: José María García / Octavio Martín-Díaz / Carlos Müller / Pablo Fernandez / Antonio Ruiz-Cortés / 
Palabras Clave: Automated analysis - customer agreements - Regulatory Compliance - Service Level Agreements

The term of API Economy is becoming increasingly used to describe the change of vision in how APIs can add value to the organizations. Furthermore, a greater automation of RESTful APIs management can suppose a competitive advantage for the company. New proposals are emerging in order to automatize some API governance tasks and increase the ease of use (e.g. generation of code and documentation). Despite that, the non-functional aspects are often addressed in a highly specific manner or even there not exists any solution for an automatic governance. Nevertheless, these properties are already defined in natural language at the Service Level Agreement (SLA) that both customer and provided have established. In this paper, we carry out a study on the *aaS industry and analyze the current both API modeling and SLA modeling proposals in order to identify the open challenges for an automatic RESTful API governance.

Autores: Antonio Gámez-Díaz / Pablo Fernandez / Antonio Ruiz-Cortés / 
Palabras Clave:

La dificultad para decidir la compra de un IaaS (Infrastructure as a Service) depende de la complejidad de las opciones de compra dadas por su proveedor y de la complejidad del plan del cliente que quiere realizarla. Es habitual que estos tipos de servicios ofrezcan muchas configuraciones de uso diferentes, y para cada una de ellas sea posible disponer de varias opciones de compra. De este modo, decidir la mejor compra se convierte en una tarea que consume mucho tiempo, tediosa y propensa a errores. En este trabajo inicial, caracterizamos el problema con un caso de estudio ilustrativo y presentamos los desafíos inmediatos para mejorar las herramientas de soporte actualmente disponibles.

Autores: Octavio Martín-Díaz / Pablo Fernández / Antonio Ruiz-Cortés / 
Palabras Clave: IaaS - Opciones de Compra - SLA

Resumen de la contribución

La aparición del paradigma de la computación en la nube ha conllevado un cambio significativo dentro de la industria de las tecnologías de la información, tanto para proveedores de servicios como para los propios consumidores. Así, existen servicios como los de Amazon Elastic Computing Cloud (EC2) o Google Compute Engine que ofrecen computación virtualizada y almacenamiento de recursos (comúmente denominados Infraestructuras como Servicios o IaaS por sus siglas en inglés), de forma que los clientes pueden adquirirlos para reducir los costes de operación de sus sistemas, en comparación con el aprovisionamiento de las mismas infraestructuras de computación en un entorno local. Sin embargo, el aprovisionamiento de servicios en la nube resulta una tarea muy compleja dada la abrumadora variedad de proveedores, configuraciones y opciones de compra disponibles. En este escenario aparecen además diversas dificultades para comparar las ofertas de los distintos proveedores, debido a la heterogeneidad en la descripción de las configuraciones, opciones de compra, o incluso descuentos aplicables. A su vez, las necesidades concretas de los consumidores podrían incluir restricciones adicionales para tener en cuenta una planificación temporal previa en cuanto al número de instancias de IaaS que necesitarán en determinados momentos. Aunque existen algunas herramientas y calculadoras on-line que permiten buscar configuraciones concretas de IaaS, éstas no tienen en cuenta cuestiones como la planificación y la optimización de las opciones de compra. En este trabajo presentamos un framework de análisis automático que es capaz de analizar y comparar ofertas de servicios en la nube de distintos proveedores para obtener un plan de aprovisionamiento óptimo de acuerdo con las necesidades de los consumidores. Dicho plan especifica la cantidad y el tipo de instancias de IaaS que deben adquirirse, junto con la planificación de su uso. Hemos desarrollado un prototipo que ha sido validado en un escenario de virtualización de clases de laboratorio, comparando las opciones de dos proveedores.

Autores: José María García / Octavio Martín-Díaz / Pablo Fernandez / Antonio Ruiz-Cortés / Miguel Toro / 
Palabras Clave:

Microservices architectures are now widely used by a great number of developers. They often consume external services offered by third parties, which usually impose limitations on the number of requests that they can serve. These limitations have an impact on the capacity of the architecture, which is determined by the number of requests that it can handle, among other factors. This paper aims to automate the capacity analysis over microservices architectures at design time, following an analytical approach. We present a generic CSOP model that can be used to analyze the capacity of any architecture. We also evaluate our proposal using a real architecture.

Autores: Rafael Fresno-Aranda / Pablo Fernández / Antonio Ruiz-Cortés / 
Palabras Clave: Architecture - automation - Capacity - Limitations - microservices

The Cloud Service Market has evolved into a complex landscape that challenges the decision making of users as they develop their purchasing process. In particular, we explore the case of cloud infrastructure (IaaS) providers as an example of heterogeneous variety of purchasing options and discounts; this variability represents an important drawback during the decision making process where there is a need to compare and select the best option. In this work, we define a common model to describe purchasing models from different providers taking into account such heterogeneity. This purchasing model represents a first step towards the automated support of decision making problems during the purchasing process. In order to illustrate our approach we apply the model in a real case study of IaaS purchasing.

Autores: Octavio Martín-Díaz / Pablo Fernandez / José María García / Antonio Ruiz-Cortés / 
Palabras Clave: Cloud Services - Decision Making - Purchasing Options

Current software industry is evolving into a service­centric scenario and consequently, the importance to create reliable service consumptions amongst organizations is a key point. In such a context, the concept of Service Level Agreement (SLA) represents the foundation to express the responsibilities (i.e. rights and obligations) of service consumer and provider during the consumption. However, in spite there has been a major effort in both academia and industry to develop languages and frameworks to support SLAs, there still remain important challenges to address such as how to automate the detection of a violation of the SLAs and how to react accordingly in order to claim for a compensation. Specifically, in this paper we focus on the definition of the automated claiming of SLAs problem characterized as the set of processes of gathering, checking and explaining the evidences associated with the service consumption within the context of an SLA. In order to identify the key requirements to automate the claiming of SLAs, we analyse the real case of the Simple Storage Service (S3) provided by Amazon, that is regulated by an SLA. Based on our analysis we propose a set of extensions to current prominent SLA language specification (WS­Agreement) and conceptualize a list of research challenges to automate the management of the claiming process.

Autores: Manuel Léon / Pablo Fernandez / Manuel Resinas / Antonio Ruiz-Cortés / 
Palabras Clave:

As distribution models of information systems are moving to XaaS paradigms, microservices architectures are rapidly emerging, having the RESTful principles as the API model of choice. In this context, the term of API Economy is being used to describe the increasing movement of the industries in order to take advantage of exposing their APIs as part of their service offering and expand its business model.

Currently, the industry is adopting standard specifications such as OpenAPI to model the APIs in a standard way following the RESTful principles; this shift has supported the proliferation of API execution platforms (API Gateways) that allow the XaaS to optimize their costs. However, from a business point of view, modeling offering plans of those APIs is mainly done ad-hoc (or in a platform-dependent way) since no standard model has been proposed. This lack of standardization hinders the creation of API governance tools in order to provide and automate the management of business models in the XaaS industry.

This work presents a systematic analysis of 69 XaaS in the industry that offer RESTful APIs as part of their business model. Specifically, we review in detail the plans that are part of the XaaS offerings that could be used as a first step to identify the requirements for the creation of an expressive governance model of realistic RESTful APIs. Additionally, we provide an open dataset in order to enable further analysis in this research line.

Autores: Antonio Gamez-Diaz / Pablo Fernandez / Antonio Ruiz-Cortés / 
Palabras Clave: API economy - RESTful API - SLA-Driven API

The number and variety of experiments carried in software engineering research is growing, leading to a increasing need of replication and review. In order to support such needs the information about experiments should be provided as lab-packs comprising of: a description of the experiment, the materials used and data generated during the conduction, and the results of the analyzes performed on such data. However, this information is often scattered, poorly structured, and even unavailable, implying a tedious process of search and gathering. EXEMPLAR is an online platform for managing experimental information, that allows the uploading and publication of experimental lab packs, and an efficient search. The platform also supports the use of formal languages for providing experimental descriptions (e.g. SEDL). In so doing, EXEMPLAR enables the automated analysis of lab-packs, in order to detect common validity threats and missing information which could hinder replicability.

Autores: Jose Antonio Parejo / Sergio Segura / Pablo Fernandez / Antonio Ruiz Cortés / 
Palabras Clave: empirical research - experimental replicability - experimental repositories - experiments