Snippet

Preventing Out-of-Memory Errors via JPA Query Result Streaming

Fetching large datasets as a single `List<T>` loads all records simultaneously into the JVM heap, risking `OutOfMemoryError`. Using Java 8 `Stream<T>` query return types alongside `@Transactional(readOnly = true)` allows the database cursor to stream rows incrementally, keeping heap consumption constant while processing extensive result sets.

snippet.java

java

@Service
public class ExportService {
 
    @Transactional(readOnly = true)
    public void exportAuditLogs(Consumer<AuditLog> consumer) {
        try (Stream<AuditLog> logStream = auditLogRepository.streamAllByStatus(Status.PENDING)) {
            logStream.forEach(consumer);
        }
    }
}

spring

Breakdown

@Transactional(readOnly = true)

Maintains the underlying database connection and cursor open throughout the stream iteration.

try (Stream<AuditLog> logStream = auditLogRepository.streamAllByStatus(Status.PENDING)) {

Opens a try-with-resources block to guarantee closing the database cursor once streaming completes.

logStream.forEach(consumer);

Processes entities sequentially as they are fetched from the cursor without buffering all entries in memory.

Previous snippet Next snippet

More Intermediate Java exercises →

Preventing Out-of-Memory Errors via JPA Query Result Streaming

Related