2021 Research Review / DAY 2
Multicore Confidence
Complex, cyber-physical DoD systems, such as aircraft, depend on correct timing to properly and reliably execute crucial sensing, computing, and actuation functions. Any timing failure can have disastrous consequences--an expected delay translating sensor data into actuation can cause system instability and loss of control. What’s more, the complexity of today’s DoD systems has increased the demand for use of multicore processors because unicore chips are either unavailable or not up to the task. However, concerns about timing have led to the practice of disabling all processor cores except one.
Any timing failure can have disastrous consequences—an expected delay translating sensor data into actuation can cause system instability and loss of control.
In this project, we aim to develop a solution to overcome this obstacle. This is a difficult challenge, because timing is determined by many shared resources in the memory system (including cache, memory banks, memory bus) with complex arbitration mechanisms, some of which are undocumented. The goal of our research is to demonstrate multicore timing confidence by achieving the following sub-objectives:
- Verification. Develop a method for timing verification that does not depend directly on undocumented design qualities and quantities.
- Parameter extraction. Develop a method for obtaining values for parameters in the model of a software system suited for the timing verification procedure mentioned above.
- Configuration. Develop a configuration procedure (such as assigning threads to processor cores or assigning priorities to threads) that takes a model as input and produces a configuration for which the verification will succeed (if such a configuration exists).

In Context:
This FY2019 project
- builds on prior DoD line-funded research and sponsored work on timing verification of undocumented multicore, verifying distributed adaptive real-time systems, high-confidence cyber-physical systems, and real-time scheduling for multicore architectures
- aligns with the CMU SEI technical objective to bring capabilities through software that make new missions possible or improve the likelihood of success of existing ones
Principal Investigator
Bjorn Andersson
Principal Researcher
SEI Collaborators
Bill Anderson
Senior Member of the Technical Staff
Dionisio de Niz
Technical Director, Assuring Cyberphysical Systems
Anton Hristozov
Software Engineer
Mark Klein
Principal Technical Advisor
External Collaborators
Hyoseung Kim
Associate Professor, Department of Electrical And Computer Engineering
University of California, Riverside
John Lehoczky
Thomas Lord University Professor of Statistics & Data Science, Department of Statistics & Data Science
Carnegie Mellon University