Computer system with moving target defenses against vulnerability attacks
AssigneeThe Regents of the University of Michigan
InventorsTodd Austin, Valeria Bertacco, Mark Gallagher, Baris Kasikci
Life of this patent
Estimating the term…
Abstract
A computer system includes an ensemble moving target defense architecture that protects the computer system against attack using a plurality of composable protection layers that change each churn cycle, thereby requiring an attacker to acquire information needed for an attack (e.g., code and pointers) and successfully deploy the attack, before the layers have changed state. Each layer may deploy a different attack information asset protection providing multiple different attack protections each churn cycle.
Claims19
A computer-implemented method of securing a computer system against unauthorized access from an unauthorized agent, the method comprising:
- assigning attack information assets, stored in the computer system and targetable by an unauthorized agent to exploit a vulnerability of the computer system, as one of a plurality of information domains consisting of a code domain, a code pointer domain, a data object domain, a data object pointer domain, a relative distance between data objects domain, and an operation timing domain;
- and
- for each information domain, protecting the attack information assets assigned to the information domain using at least one of a plurality of composable layers of protection and executable during runtime of the computer system,
- wherein each composable layer of protection is configured to provide a different attack information asset protection, and
- wherein at least one of the plurality of composable layers is a moving target defense layer having a churn cycle and a churn rate wherein the attack information asset protection of the moving target defense layer is reconfigured each churn cycle at the churn rate.
A computer-implemented method of securing a computer system against unauthorized access from an unauthorized agent, the method comprising:
- assigning attack information assets, stored in the computer system and targetable by an unauthorized agent to exploit a vulnerability of the computer system, as corresponding to one of a plurality of information domains consisting of a code domain, a code pointer domain, a data object domain, a data object pointer domain, a relative distance between data objects domain, and an execution timing domain;
- for each information domain, protecting the assigned attack information assets using a plurality of different composable layers of protection each hardware stored in the computer system and executable during runtime of the computer system, and
- wherein each composable layer of protection is configured to provide a different attack information asset protection, and wherein at least one of the plurality of composable layers is a reconfigurable composable layer.
A computer-implemented method of securing a computer system against unauthorized access from an unauthorized agent, the method comprising:
- identifying attack information assets targetable by an unauthorized agent to exploit a vulnerability of the computer system as corresponding to one of a plurality of information domains consisting of a code domain, a code pointer domain, a data object domain, a data object pointer domain, a relative distance between data objects domain, and an execution timing domain;
- defining a plurality of composable layers of protection each configured to provide a different attack information asset protection;
- ensembling the plurality of composable layers to simultaneously protect the attack information assets against vulnerability of the computer system;
- reconfiguring the ensembled plurality of composable layers at a churn rate such that the vulnerability of the computer system is changed each churn cycle;
- and
- layering the ensemble plurality of composable layers such as at least two composable layers protect the same information domain.
Dependencies are inferred from the claim text.