Álvaro Navas

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 different 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.

A Multi-Level Monitoring Approach for the Dynamic Management of Private IaaS Platforms

To properly implement a dynamic management system in a cloud environment it is imperative to have complete, holistic and up to date information of the environment itself. However, current private cloud IaaS monitoring solutions are not sufficiently evolved to fulfill this task. Such a monitoring system needs to contemplate all stakeholders and dimensions of a cloud environment, while at the same time hiding irrelevant information. We propose an information model that offers a multi-level view of a private IaaS cloud, separates the concerns between consumers and providers, and enables the correlation of the collected data to have a complete picture of the environment. Using it we have designed a monitoring architecture, and built it using agents based on open-source monitoring platforms. We show the feasibility of our approach through two sample scenarios deployed into a private IaaS cloud. This monitoring provides the capabilities needed to be used in an autonomous management system.

Research challenges for cross-cloud applications

Federated clouds can expose the Internet as a homogeneous compute fabric. There is an opportunity for developing cross-cloud applications that can be deployed pervasively over the Internet, dynamically adapting their internal topology to their needs. In this paper we explore the main challenges for fully realizing the potential of cross-cloud applications. First, we focus on the networking dimension of these applications. We evaluate what support is needed from the infrastructure, and what are the further implications of opening the networking side. On a second part, we examine the impact of a distributed deployment for applications, assessing the implications from a management perspective, and how it affects the delivery of quality of service and non-functional requirements.

A Change Execution System for Enterprise Services with Compensation Support

Modern enterprises rely on a distributed IT infrastructure to execute their business processes, adopting Service Oriented Architectures in order to improve the flexibility and ease of adaptation of their functions. Nowadays this is a vital characteristic, as the increased competition forces companies to continuously evolve and adapt. SOA applications must be supported by management and deployment systems, which have to continuously apply modifications to the distributed infrastructure. This article presents a model-based solution for automatically applying change plans to heterogeneous enterprise managed environments. The proposed solution uses models which describe in an abstract language the changes that need to be applied to the environment, and executes all the required operations to the specific managed elements. Also, to ensure that the environment ends in a stable state, compensation for previously executed operations is supported. The validation results from a case study taken from the banking domain are also presented here.

A Tool for the Analysis of Change Management Processes in Software Development Cycles

Change management processes theory specifies the life cycle of a change through an organization. It is a wellknown process present in day-to-day operations, with up to hundreds of changes passing through its phases each day. There is a broad range of tools that help with keeping track of each of those changes. However, the use of these tools, and hence the process itself, is not always translated perfectly into an organization. Therefore, it is necessary to analyse how the process has been implemented and how to correct it. Change management systems often offer some degree of analysis, but it is either too little or too obtuse. In this paper we present a tool that can help analyse the data gathered by these systems in order to detect bottle-necks and irregularities in a visual way tailored to the special time needs of the data.

REST service testing based on inferred XML schemas

The concept of service oriented architecture has been extensively explored in software engineering, due to the fact that it produces architectures made up of several interconnected modules, easy to reuse when building new systems. This approach to design would be impossible without interconnection mechanisms such as REST (Representationa State Transfer) services, which allow module communication while minimizing coupling. . However, this low coupling brings disadvantages, such as the lack of transparency, which makes it difficult to sistematically create tests without knowledge of the inner working of a system. In this article, we present an automatic error detection system for REST services, based on a statistical analysis over responses produced at multiple service invocations. Thus, a service can be systematically tested without knowing its full specification. The method can find errors in REST services which could not be identified by means of traditional testing methods, and provides limited testing coverage for services whose response format is unknown. It can be also useful as a complement to other testing mechanisms.

A Repository for Integration of Software Artifacts with Dependency Resolution and Federation Support

While developing new IT products, reusability of existing components is a key aspect that can considerably improve the success rate. This fact has become even more important with the rise of the open source paradigm. However, integrating different products and technologies is not always an easy task. Different communities employ different standards and tools, and most times is not clear which dependencies a particular piece of software has. This is exacerbated by the transitive nature of these dependencies, making component integration a complicated affair. To help reducing this complexity we propose a model-based repository, capable of automatically resolve the required dependencies. This repository needs to be expandable, so new constraints can be analyzed, and also have federation support, for the integration with other sources of artifacts. The solution we propose achieves these working with OSGi components and using OSGi itself.