predictive mitigation of timing channels - threat defense for machine codes

The security of software systems can be threatened by many internal and external threats, including data leakages due to timing channels. even if developers manage to avoid security threats in the source code or bytecode during development and testing, new threats can arise as the compiler generates machine codes from representations at the binary code level during execution on the processor or due to operating system specifics. current approaches either do not allow the neutralization of timing channels to be achieved comprehensively with a sufficient degree of security or require an unjustifiable amount of time and/or resources. herein, a method is demonstrated for the protected execution of software based on a secure virtual execution environment (vee) that combines the results from dynamic and static analyses to find timing channels through the application of code transformations. this solution complements other available techniques to prevent timing channels from being exploited. this approach helps control the appearance and neutralization of timing channels via just-in-time code modifications during all stages of program development and usage. this work demonstrates the identification of threats using timing channels as an example. the approach presented herein can be expanded to the neutralization of other types of threats.