Section: New Results

Models for reliability

Participants : Mario Bravetti, Elena Giachino, Ivan Lanese, Michael Lienhardt, Gianluigi Zavattaro.

Reversibility

We have continued the study of causal-consistent reversibility started in the past years. In [17] we presented an overview of causal-consistent reversibility, summarizing the main approaches in the literature, and the related results and applications. An interesting application is debugging. Reversible debugging provides developers with a way to execute their applications both forward and backward, seeking the cause of a misbehaviour. In a concurrent setting, reversing actions in the exact reverse order they have been executed may lead to undo many actions that were not related to the bug under analysis. On the other hand, undoing actions in some order that violates causal dependencies may lead to states that could not be reached in a forward execution. In [36] we proposed a new approach, where each action can be reversed if all its consequences have already been reversed. The main feature of the approach is that it allows the programmer to easily individuate and undo exactly the actions that caused a given misbehaviour till the corresponding bug is reached. We discussed the appropriate primitives for causal-consistent reversible debugging and presented their prototype implementation in the CaReDeb tool.

Fault models

We have continued the study of primitives for fault handling in a concurrent setting. In [19] we critically discussed the different choices that have to be made when defining a fault model for a concurrent object-oriented programming language. We consider in particular the ABS language, and analyse the interplay between the fault model and the main features of ABS, namely the cooperative concurrency model, based on asynchronous method invocations whose return results via futures, and its emphasis on static analysis based on invariants.