Introduction
Modern societies critical rely on numerous software-centric systems and services for the majority of their core services spanning the composite of critical infrastructure control & management (power, water, etc.); eCommerce & M-Commerce; finance & banking; business-to-business systems; eGovernment & eVoting; advanced health care & eHealth; social media & entertainment; etc. Societies reliance means these systems must be engineered to be safe, secure, protect private and proprietary information, reliable, available, performance, adaptable, maintainable, etc. Emerging domains such as Smart Cities, Smart Grids, autonomous vehicles and drones will further accelerate this reliance while adding even more immediate human and environmental health and safety concerns.
The Information Security and Privacy (InSPiRe) lab focus on advancing the required formal engineering within these areas such that we can build industry-scale operational systems that behave predictably and deliver on societies’ expectations. An important and critical aspects of the InSPiRe lab and the research we do is the deep and active connections within industry and industry-scale systems and needs, inclusive of active engagement in high-tech entrepreneurship and the confounding of a number of successful, globally competitive companies.
Research questions and issues the InSPiRe lab has and is engaged in include:
- When, where, and why do modern cloud-deployed large-scale software systems exhibit (or not exhibit) statistically predictable run-time behaviors?
- This is a critical and foundational issue within a wide set of areas including the optimization of cloud platforms, the optimization of hybrid software system architectures, the security of cloud systems, the “greening” of cloud computing, etc.
- When and why do composite sets of industry best-practice security solutions lead to bounded security risk against capable intelligent adversaries?
- The nature of such questions remains open and challenging as per the volumes of occurring daily security occur daily which serve to highlight that current security methodologies are necessary but not sufficient to bound risk.
- When and why do modern large-scale machine learning and AI methodologies and approaches fail within operational scale industry contexts?
- Current reports are highlighting that 70% to 80% of production-scale AI/ML solutions are failing in industry, even though the methodologies on which they are based have been shown to work well within far more constrained academic research lab setting and environments.
Common across these questions and similar ones of interest within InSPiRe are the need to correctly, properly, and completely understand the behaviors and complexities exhibited by real-world industry scale operational systems. As such, InSPiRe has deep and active industry engagement and collaborations, including in high-tech entrepreneurship and the creation and co-founding, to date, of seven successful and globally competitive high-tech companies.
From a theoretical perspective the InSPiRe lab focused on the application of and mathematical insights provided via areas such as:
- Statistical data analysis, pattern recognition, and signal processing (time series analyses)
- Machine learning and AI
- Dynamical systems theory and Ergodic theory
- Game theory (as per the actions of intelligent adversaries)
- Optimization theory
As such InSPiRe is strongly focused on the engineering of systems and not the far less onerous task of simply building them, i.e., as per modern AI/ML many systems can be easily built but very few are then working correctly and as intended at industry production-scales.
Supporting this research is a purpose-build cluster-computing test bed facility dedicated to reproducing enterprise-scale network environments while meeting the full tenets of scientific control and repeatability.
This facility consists of:
- $500k+ in hardware
- an additional $250k+ investment in the R&D of its associated control/management software system.
© 2003 – 2025 < Stephen W. Neville >