arXiv subjects

Victor Vianu

Publications and source records attributed to Victor Vianu.

4 recordsLinked to original sources

arXiv2018

SpinArt: A Spin-based Verifier for Artifact Systems

Data-driven workflows, of which IBM's Business Artifacts are a prime exponent, have been successfully deployed in practice, adopted in industrial standards, and have spawned a rich body of research in academia, focused primarily on static analysis. In previous work, we obtained theoretical results on the verification of a rich model incorporating core elements of IBM's successful Guard-Stage-Milestone (GSM) artifact model. The results showed decidability of verification of temporal properties of a large class of GSM workflows and established its complexity. Following up on these results, the present paper reports on the implementation of SpinArt, a practical verifier based on the classical model-checking tool Spin. The implementation includes nontrivial optimizations and achieves good performance on real-world business process examples. Our results shed light on the capabilities and limitations of off-the-shelf verifiers in the context of data-driven workflows.

cs.DB↗

arXiv2017

VERIFAS: A Practical Verifier for Artifact Systems

Data-driven workflows, of which IBM's Business Artifacts are a prime exponent, have been successfully deployed in practice, adopted in industrial standards, and have spawned a rich body of research in academia, focused primarily on static analysis. The present research bridges the gap between the theory and practice of artifact verification with VERIFAS, the first implementation of practical significance of an artifact verifier with full support for unbounded data. VERIFAS verifies within seconds linear-time temporal properties over real-world and synthetic workflows of complexity in the range recommended by software engineering practice. Compared to our previous implementation based on the widely-used Spin model checker, VERIFAS not only supports a model with richer data manipulations but also outperforms it by over an order of magnitude. VERIFAS' good performance is due to a novel symbolic representation approach and a family of specialized optimizations.

cs.DB↗

arXiv2017

Research Directions for Principles of Data Management (Dagstuhl Perspectives Workshop 16151)

In April 2016, a community of researchers working in the area of Principles of Data Management (PDM) joined in a workshop at the Dagstuhl Castle in Germany. The workshop was organized jointly by the Executive Committee of the ACM Symposium on Principles of Database Systems (PODS) and the Council of the International Conference on Database Theory (ICDT). The mission of this workshop was to identify and explore some of the most important research directions that have high relevance to society and to Computer Science today, and where the PDM community has the potential to make significant contributions. This report describes the family of research directions that the workshop focused on from three perspectives: potential practical relevance, results already obtained, and research questions that appear surmountable in the short and medium term.

cs.DB↗

arXiv2016

Verification of Hierarchical Artifact Systems

Data-driven workflows, of which IBM's Business Artifacts are a prime exponent, have been successfully deployed in practice, adopted in industrial standards, and have spawned a rich body of research in academia, focused primarily on static analysis. The present work represents a significant advance on the problem of artifact verification, by considering a much richer and more realistic model than in previous work, incorporating core elements of IBM's successful Guard-Stage-Milestone model. In particular, the model features task hierarchy, concurrency, and richer artifact data. It also allows database key and foreign key dependencies, as well as arithmetic constraints. The results show decidability of verification and establish its complexity, making use of novel techniques including a hierarchy of Vector Addition Systems and a variant of quantifier elimination tailored to our context.

cs.DB↗