In Guidewire InsuranceSuite, managing how data is retrieved from the database is critical for system performance, specifically regarding memory usage and the bundle size. A primary goal for any developer is to ensure that filtering happens at the database level rather than within the application server ' s memory.
Options B and C demonstrate a common but inefficient pattern: accessing an array directly (e.g., claim.
Activities). When you access an array on an entity, the Guidewire platform automatically loads every related record in that array into the application server ' s memory and adds them to the current Bundle. If a claim has hundreds of activities, but you only need the three that are " Urgent " and " Open, " Options B and C still force the system to load all of them. This consumes significant memory and increases the overhead of the bundle, which can lead to performance degradation or " Out of Memory " errors in high-volume environments.
Option D is the verified best practice. By using the Gosu Query API (Query.make(Activity)), the developer can build a specific SQL statement. Using the .compare() method for the Priority, Status, and the link to the Claim ensures that all three criteria are passed to the database as part of the WHERE clause. When .select() is called, the database engine filters the records and returns only the specific rows that meet all requirements.
Consequently, only the necessary objects are loaded into the application server ' s memory and added to the bundle. Option A is less efficient because it uses a Gosu lambda (.where) after the select, which performs the final filtering in memory rather than at the database tier. Following the pattern in Option D minimizes the data
" payload " and ensures the application remains scalable and responsive.