"After years of ignoring some key JPA methods and annotations, I finally explored them - and what happened next was shocking: my application became up to 10× faster with fewer queries, less memory use, and way cleaner code. 🔄 1. JpaRepository.existsById() Instead of findById() If you only need to check if something exists: if (userRepository.existsById(id)) { // no need to load entire entity} ✅ Why it's faster: Avoids unnecessary entity loading and reduces SQL size."
"my application became up to 10× faster with fewer queries, less memory use, and way cleaner code. 💾2. saveAll() - The Implicit Batch Saver For years, I called save() in a loop - like this: for (User user : users) { userRepository.save(user);} But that's inefficient. Instead, use: userRepository.saveAll(users); Hint: saveAll() still calls save() for each entity internally, but the persistence context can optimize batching if you configure it correctly (e.g., using hibernate.jdbc.batch_size)."
Using JPA methods like existsById avoids loading full entities when only existence checks are needed, reducing SQL size and memory use. Calling saveAll instead of save in a loop enables the persistence context and JDBC batching to reduce database roundtrips and queries when configured (for example, hibernate.jdbc.batch_size). saveAll still invokes save for each entity, but batching and the persistence context optimize operations and memory. These changes can produce significantly fewer queries, lower memory consumption, cleaner code, and performance improvements of up to ten times in some cases.
Read at Medium
Unable to calculate read time
Collection
[
|
...
]