Design of Approaches for Dependability and Initial Prototypes

Antonia Bertolino, Antonello Calabrò, Silvano Chiaradonna, Gabriele Costa, Félicita Di Giandomenico, Antinisca Di Marco, Mario Fusani, Fabrizio Grandoni, Valérie Issarny, Marta Kwiatkowska, Eda Marcheti, Fabio Martinelli, Marco Martinucci, Paolo Masci, Ilaria Matteucci, Hongyang Qu, Rachid Saadi, Antonino Sabetta, Anna VaccarelliHAL 2011

TL;DR

This paper introduces the CONNECTability framework, a comprehensive approach to ensuring non-functional properties like dependability, performance, security, and trust in interconnected systems. It presents novel methods for analyzing these properties at both design and runtime, paving the way for more robust and interoperable networked systems.

What Problem Does This Paper Address?

The paper tackles the challenge of achieving universal interoperability between diverse networked systems. As systems become increasingly interconnected, ensuring that they not only function correctly but also meet critical non-functional requirements (dependability, performance, security, and trust) becomes paramount. Traditional approaches often struggle with the dynamic and evolving nature of these systems, leading to potential vulnerabilities and performance issues. The CONNECT project aims to address this by developing a framework that can analyze and guarantee these properties, both during the design phase and continuously at runtime.

Methods

The paper presents several key contributions:

  • CONNECTability Framework: A unified concept and meta-model (CPMM) for defining and expressing non-functional properties, making them machine-processable for automated analysis and synthesis.
  • Dependability & Performance Analysis Enabler (DePer): A modular architecture supporting both stochastic model checking and state-based stochastic methods for analyzing dependability and performance. This includes prototypes based on PRISM and Möbius.
  • Incremental Verification and Runtime Analysis: Techniques to refine dependability analysis over time by reusing previous results and incorporating runtime data. This includes an SCC-based approach to optimize model checking and an integration with a monitoring enabler (GLIMPSE).
  • Interoperable Trust Management: A formal trust meta-model and composition mechanisms to handle heterogeneous trust models across different networked systems.
  • Security-by-Contract-with-Trust (SxCxT): A framework that integrates security and trust for runtime enforcement of policies and contracts, with a prototype implemented in Java Micro Edition.

Datasets

Not discussed in the paper. The paper focuses on developing methodologies and frameworks, and uses a specific scenario (Terrorist Alert) for demonstration rather than training on large datasets.

Key Insights and Ideas

  • Unified Concept of "CONNECTability": By introducing "CONNECTability," the authors provide a single, overarching term to encompass dependability, performance, security, and trust, simplifying the discussion and integration of these crucial non-functional aspects.
  • Runtime Analysis is Crucial for Evolving Systems: The paper strongly emphasizes the need for continuous analysis and refinement of system properties at runtime, acknowledging that static analysis alone is insufficient for dynamic and evolving networked environments.
  • Trust and Security are Intertwined: The work highlights the critical interplay between trust and security, demonstrating how trust levels can influence the enforcement of security policies and vice versa.
  • Model-Driven Approach for Non-Functional Properties: The development of a formal meta-model (CPMM) and associated tools (EMF editor) enables a model-driven approach to specifying and analyzing non-functional properties, promoting consistency and automation.
  • Complementarity of Analysis Techniques: The paper showcases how different analysis techniques (stochastic model checking and state-based stochastic methods) can complement each other, offering a more comprehensive assessment of system properties.

Key Results

  • Demonstrated Frameworks: The paper presents working prototypes for the Dependability&Performance Analysis Enabler, incremental verification, and the Security-by-Contract-with-Trust mechanism.
  • Scenario Validation: The proposed approaches were successfully applied to the "Terrorist Alert" scenario, demonstrating their ability to model and analyze properties like latency, coverage, and privacy.
  • Performance Improvements: Incremental verification techniques showed significant speed-ups (up to 50x) compared to non-incremental methods, making runtime analysis more feasible.
  • Scalability: The Möbius-based approach demonstrated scalability by analyzing scenarios with hundreds of guards.

Limitations and Future Work

  • Trust Model Refinement: The trust meta-model is still under development, with further work planned for its completion and integration.
  • Complex Event Specification: The specification of complex events in the EventType meta-class needs further refinement.
  • Action-Based Expression Semantics: The definition of action-based expressions for trust modeling requires further development and validation.
  • State-Based Properties: The current meta-model primarily supports action-based properties; extending it to handle state-based properties is a future direction.
  • Ontology Integration: Integrating CPMM with ontologies for non-functional requirements is planned to introduce richer semantics.
  • Model-to-Model and Model-to-Code Transformations: Developing automated transformations to translate property models into specific notations for analysis and monitoring tools is a key future task.
  • Full Integration of Enablers: A significant focus for the third year is the integration and convergence of WP5 Enablers with other CONNECT Enablers.
  • Cryptography in Security: Extending the enforcement strategy to handle cryptography is a future research direction.
  • Automated Trust Negotiation: Further work on automated trust negotiation among NSs based on soft constraints is planned.

Summary updated

Abstract

The aim of CONNECT is to achieve universal interoperability between heterogeneous Networked Systems. For this, the non-functional properties required at each side of the connection going to be established must be fulfilled. By the one inclusive term "CONNECTability" we comprehend properties belonging to all four non-functional concerns of interest for CONNECT, namely dependability, performance, security and trust. We model such properties in conformance with a meta-model which establishes the relevant concepts and their relations. Then, building on the conceptual models proposed in the first year in Deliverable D5.1, in this document we present the approaches developed for assuring CONNECTability both at synthesis time and at runtime. The contributions include: the Dependability&Performance analysis Enabler, for which we release a modular architecture supporting stochastic verification and state-based analysis; incremental verification and event-based monitoring for runtime analysis; a model-based approach to interoperable trust management; the Security-by-Contract-with-Trust framework, which guarantees and enforces the expected trust levels and security policies.
BibTeX
@inproceedings{bertolino2011design,
  author = {Antonia Bertolino and Antonello Calabrò and Silvano Chiaradonna and Gabriele Costa and Félicita Di Giandomenico and Antinisca Di Marco and Mario Fusani and Fabrizio Grandoni and Valérie Issarny and Marta Kwiatkowska and Eda Marcheti and Fabio Martinelli and Marco Martinucci and Paolo Masci and Ilaria Matteucci and Hongyang Qu and Rachid Saadi and Antonino Sabetta and Anna Vaccarelli},
  title = {{Design of Approaches for Dependability and Initial Prototypes}},
  booktitle = {HAL},
  year = {2011},
}
Figure