CritiX

The Critical and Extreme Security and Dependability Research Group - CritiX - is headed by Professor Marcus Völp. CritiX has been founded by Paulo Esteves-Veríssimo and pursues state-of-the-art research in a problem area that may be described as extreme computing -- computer science and engineering pushed to the extremes of functional and non-functional properties of systems. In particular, we wish to investigate architectures, middleware, algorithms and protocols that may find applicability in distributed systems and networks, which, for example:

• Deploy extremely large-scale data sets, flows and computations --- considering cloud, BigData, complex event processing
• Withstand extreme levels of threat, like advanced persistent threats --- considering critical information infrastructures
• Need to have extremely low failure probability --- considering high-criticality areas such as finance, energy, networking (SDN), or aerospace and autonomous vehicles
• Present extreme requirements with regard to data privacy and integrity ---considering e-health, genomics, or business/finance

Resilient Modular and Distributed Computing is our response to the need for a paradigm shift enabling a comprehensive approach to those extreme challenges, from first principles: architecting and designing for simultaneously coping with accidental and malicious faults; providing protection in an incremental way; and automatically adapting to a dynamic range of scale, severity and persistence of threats, some of which maybe a priori unknown. Paradigms and techniques emerging from this research should endow systems with the capacity of defeating extreme adversarial power, accidental or malicious (severe and continued threats) and sustaining perpetual and unattended operation (in a systematic and automatic way).

We plan on addressing this level of threat drawing from and building on recent research on powerful and innovative automatic security and dependability techniques, like fault and intrusion tolerance or Byzantine fault tolerance (BFT), trusted computing and architectural hybridisation, secret sharing and secure multi-party computation, self-healing and diversity, or post-compromise security. Furthermore, we will leverage enhanced formal verification techniques such as interactive theorem proving, to achieve ultra-high reliance on software used behind roots-of-trust or TCBs.

We would like our laboratory to become known as excelling in research applicable to systems that face difficult or extreme situations (of scale, threat, environmental, operational, uncertainty, and so forth).