Paulo Jesus

During software testing, defect prediction approaches measure current reliability status, forecasting future program failures, and provide information on how many defects need to be removed before shipping. Existing approaches often require faults to be detected and identified as a new one, before a model-based trend can be fitted. While during regression testing failures may frequently occur, it is not evident which are related to new faults. Consequently, reliability growth trending can only be performed in sync with fault identification and repair, which is often performed in between regression test cycles. In this paper we present a dynamic, reasoning approach to estimate the number of defects in the system early in the process of regression testing. Our approach, coined Dracon, is based on Bayesian fault diagnosis over abstractions of program traces (also known as program spectra). Experimental results show that Dracon systematically estimates the exact number of (injected) defects, provided sufficient tests cases are available. Furthermore, we also propose a simple, analytic performance model to assess the influence of failed test cases in the estimation. We observe that our empirical findings are in agreement with the model.

Gossip, or epidemic, protocols have emerged as a powerful strategy to implement highly scalable and resilient reliable broadcast primitives on large scale peer-to-peer networks. Epidemic protocols are scalable because they distribute the load among all nodes in the system and resilient because they have an intrinsic level of redundancy that masks node and network failures. This chapter provides an introduction to gossip-based broadcast on large-scale unstructured peer-to-peer overlay networks: it surveys the main results in the field, discusses techniques to build and maintain the overlays that support efficient dissemination strategies, and provides an in-depth discussion and experimental evaluation of two concrete protocols, named HyParView and Plumtree.

Motivated by the need to formalize generation of fast running code for linear algebra applications, we show how an index-free, calculational approach to matrix algebra can be developed by regarding matrices as morphisms of a category with biproducts. This shifts the traditional view of matrices as indexed structures to a type-level perspective analogous to that of the pointfree algebra of programming.The derivation of fusion, cancellation and abide laws from the biproduct equations makes it easy to calculate algorithms implementing matrixmultiplication, the kernel operation of matrix algebra, ranging from its divide-and-conquer version to the conventional, iterative one. From errant attempts to learn how particular products and coproducts emerge from biproducts, we not only rediscovered block-wise matrix combinators but also found a way of addressing other operations calculationally such as e.g. Gaussian elimination. A strategy for addressing vectorization along the same lines is also given.

Puzzles and games have been used for centuries to nurture problem-solving skills. Although often presented as isolated brain-teasers, the desire to know how to win makes games ideal examples for teaching algorithmic problem solving. With this in mind, this paper explores one-person solitaire-like games.
The key to understanding solutions to solitaire-like games is the identification of invariant properties of polynomial arithmetic. We demonstrate this via three case studies: solitaire itself, tiling problems and a collection of novel one-person games. The known classification of states of the game of (peg) solitaire into 16 equivalence classes is used to introduce the relevance of polynomial arithmetic. Then we give a novel algebraic formulation of the solution to a class of tiling problems. Finally, we introduce an infinite class of challenging one-person games inspired by earlier work by Chen and Backhouse on the relation between cyclotomic polynomials and generalisations of the seven-trees-in-one type isomorphism. We show how to derive algorithms to solve these games.

Abstract. Intensive research and development has been conducted in the design and creation of groupware systems for distributed users. While for some activities, these groupware tools are widely used, for other activities the impact in the groupware community has been smaller and can be improved. One reason for this fact is that the mostly common used applications do not support collaborative features and users are reluctant to change to a different application. In this paper we discuss how available file system mechanisms can help to address this problem. In this context, we present Forby, a system that allows to provide groupware features to distributed users by combining filesystem monitoring and distributed event dissemination. To demonstrate our solution, we present three systems that rely on Forby for providing groupware features to users running unmodified applications.

Communication of large data volumes is a core functionality of distributed systems middleware, namely, for interconnecting components, for distributed computation and for fault tolerance. This common functionality is how- ever achieved in different middleware platforms with various combinations of operating system and application level protocols, both standardized and ad hoc, and including implementations on managed runtime environments such as Java.
In this paper, in contrast with most previous work that focus on performance, we point out that architectural and implementation decisions have an impact in throughput stability when the system is heavily loaded, precisely when such sta- bility is most important. In detail, we present an experimental evaluation of several communication protocol components under stress conditions and conclude on the relative merits of several architectural options.

Current model-based approaches to software debugging use static program analysis to derive a model of the program. In contrast, in the software engineering domain diagnosis approaches are based on analyzing dynamic execution behavior. We present a model-based approach where the program model is derived from dynamic execution behavior, and evaluate its diagnostic performance on the Siemens software benchmark, extended by us to accommodate multiple faults. We show that our approach outperforms other model-based software debugging techniques, which is partly due to the use of De Kleer’s intermittency model to account for the variability of software component behavior.