Debugging support for multi-paradigm concurrent programs
Dominik Aumayr
Abstract
Dominik Aumayr
Abstract
With the widespread adoption of concurrent programming, debugging of non-deterministic failures becomes increasingly important. Record & replay debugging aids developers in this effort by reliably reproducing recorded bugs. Because each concurrency model (e.g., threads vs actors) is particularly suited for different tasks, developers started combining them within the same application. Record & replay solutions are typically designed for one concurrency model only. In this paper we propose a novel multi-paradigm record & replay that is based on abstracting concurrency models to a common set of concepts and events.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
With the widespread adoption of concurrent programming, debugging of non-deterministic failures becomes increasingly important. Record & replay debugging aids developers in this effort by reliably reproducing recorded bugs. Because each concurrency model (e.g., threads vs actors) is particularly suited for different tasks, developers started combining them within the same application. Record & replay solutions are typically designed for one concurrency model only. In this paper we propose a novel multi-paradigm record & replay that is based on abstracting concurrency models to a common set of concepts and events.
Key concepts: Debugging, Concurrency, Computer science, Programming language, Set (abstract data type), Concurrent computing, Concurrent object-oriented programming, Communicating sequential processes