eResearch NZ
Browse

Integrating information flow control within programming languages and operating systems to effect data sovereignty tracking

Download (215.81 kB)
presentation
posted on 2025-03-05, 05:15 authored by eRNZ AdmineRNZ Admin, David Eyers

Information flow control (IFC) is a computer security approach that involves attaching labels to sensitive data and tracking these labels as they pass through the transformations applied by software systems.  Within programming languages and operating systems, IFC can be used to enforce mandatory access control (MAC), where participants (e.g., users and applications) within the software system also carry labels, and various security rules apply by default, based on comparisons between the IFC labels of participants and data items. 

This type of security is commonly applied within military and government intelligence sectors, with rules such as “no write down”, that indicates information cannot be declassified without an explicit permission to do so, and “no read up” that indicates that participants without permission cannot read data that is labelled as being more sensitive than their own level.  However most computer security today employs an alternative to MAC, namely discretionary access control (DAC), where the creator of a data item can arbitrary change its permissions.

In this work we present early-stage experimentation with IFC technologies for carrying labels that support data sovereignty constraints, e.g., for the support of Maori data sovereignty.  While most mainstream software systems, such as relational databases, have little or no support for IFC protections, there is a growing need to systematically handle sensitive data throughout its lifecycle, and IFC provides a promising means for achieving secure data handling within OS and programming language contexts.  Here we focus more on the mechanisms by which labels are tracked and rules are specified to use the labels for protection, with a view to how these low-level protection mechanisms might support interaction with emerging technology such as data spaces. 

ABOUT THE AUTHOR

David Eyers is a professor in the School of Computing at the University of Otago.  He has broad research interests covering security, AI, cloud, operating systems, networks and distributed systems.  One of his research themes involves investigating how emerging, low-level computer security technologies can support high-level security goals that are meaningful to the users of these computer systems.


-----

For more information about the eResearch NZ / eRangahau Aotearoa conference, visit:
https://eresearchnz.co.nz/


History

Usage metrics

    eResearch NZ

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC