This project offers a Python-based utility for evaluating the serialisability of execution schedules within database systems. It comprises two essential components: schedule.py and serialisabilitytester.py. This README provides an in-depth understanding of these components and their usage.
schedule.py serves as the foundation for this project, providing classes and structures to model and manipulate execution schedules and database operations. Key elements include:
-
IOOperation: This class represents an individual read or write operation on a database. It encapsulates attributes such as transaction ID, operation type (read or write), and the specific database element being accessed. -
Schedule: This class defines an ordered collection ofIOOperations. It captures the chronological sequence of read and write operations that occur within a database execution schedule. -
SimpleSchedule: This class represents an ordered list of transaction IDs (integers). It signifies the order in which transactions are executed in a serial fashion.
These components provide the necessary data structures for defining and analyzing execution schedules.
serialisabilitytester.py implements critical functionality for assessing the serialisability of execution schedules. Key features include:
-
to_serial(schedule): This method accepts aScheduleas input and returns the lexicographically smallest serialSimpleSchedulethat is conflict-equivalent to the provided schedule. If the input schedule is not conflict-serialisable, it returnsNone. -
conflicts: This script identifies conflicts between transactions based on read and write operations on the same database element. It then employs a topological sort to determine a serial order of execution.
To effectively utilize this project, follow these steps:
-
Clone the repository to your local machine:
git clone https://github.com/your-username/serialisability-tester.git