By leveraging large clusters of commodity hardware, the Cloud offers great opportunities to optimize the operative costs of software systems, but impacts significantly on the reliability of software applications. The lack of control of applications over Cloud execution environments largely limits the applicability of state-of-the-art approaches that address reliability issues by relying on heavyweight training with injected faults.In this paper, we propose LOUD, a lightweight fault localization approach that relies on positive training only, and can thus operate within the constraints of Cloud systems. LOUD relies on machine learning and graph theory. It trains machine learning models with correct executions only, and compensates the inaccuracy that derives from training with positive samples, by elaborating the outcome of machine learning techniques with graph theory algorithms. The experimental results reported in this paper confirm that LOUD can localize faults with high precision, by relying only on a lightweight positive training.

The charged Banados-Teitelboim-Zanelli (BTZ) black hole is plagued by several pathologies: a) Presence of divergent boundary terms in the action, hence of a divergent black hole mass; b) Once a finite, renormalized, mass M is defined black hole states exist for arbitrarily negative values of M; c) There is no upper bound on the charge Q. We show that these pathological features are an artifact of the renormalization procedure. They can be completely removed by using an alternative renormalization scheme leading to a different definition M_0 of the black hole mass, which is the total energy inside the horizon. The new mass satisfies a BPS-like bound M_0\ge (\pi/2)Q^2 and the heat capacity of the hole is positive. We also discuss the black hole thermodynamics that arises when M_0 is interpreted as the internal energy of the system. We show, using three independent approaches (black hole thermodynamics, Einstein equations, Euclidean action formulation) that M_0 satisfies the first law if a term describing the mechanical work done by the electrostatic pressure is introduced.Comment: Two references and a footnote adde

String-brane interactions provide an ideal framework to study the dynamics of the massive states of the string spectrum in a non-trivial background. We present here an analysis of tree-level amplitudes for processes in which an NS-NS string state from the leading Regge trajectory scatters from a D-brane into another state from the leading Regge trajectory, in general of a different mass, at high energies and small scattering angles. This is done by using world-sheet OPE methods and effective vertex operators. We find that this class of processes has a universal dependence on the energy of the projectile. We then compare the result for these inelastic processes with that which one would obtain from the eikonal operator in a non-trivial test of its ability to describe transitions between different string mass levels. The two are found to be in agreement.arXiv:1107.4321v2 [hep-th]

This paper presents a case study of a large software system, Netbeans 6.0, made of independent subsystems, which are analyzed as complex software networks. Starting from the source code we built the associated software graphs, where classes represent graph nodes and inter-class relationships represent graph edges. We computed various metrics for the software systems and found interdependences with various quantities computed by mean of the complex network analysis. In particular we found that the number of communities in which the software networks can be partitioned and their modularity, average path length and mean degree can be related to the amount of bugs detected in the system. This result can be useful to provide indications about the fault proneness of software clusters in terms of quantities related to the associated graph structure

We present a study of 600 Java software networks with the aim of characterizing the relationship among their defectiveness and community metrics. We analyze the community structure of such networks, defined as their topological division into subnetworks of densely connected nodes. A high density of connections represents a higher level of cooperation between classes, so a well-defined division in communities could indicate that the software system has been designed in a modular fashion and all its functionalities are well separated. We show how the community structure can be an indicator of well-written, high quality code by retrieving the communities of the analyzed systems and by ranking their division in communities through the built-in metric called modularity. We found that the software systems with highest modularity possess the majority of bugs, and tested whether this result is related to some confounding effect. We found two power laws relating the maximum defect density with two different metrics: the number of detected communities inside a software network and the clustering coefficient. We finally found a linear correlation between clustering coefficient and number of communities. Our results can be used to make predictive hypotheses about software defectiveness of future releases of the analyzed systems.