hands-on coding · Hybrid cohort
JPA + Hibernate Tuning Under Load
Batch sizing, fetch graphs, and second-level cache decisions without cargo-cult annotations.
- Duration
- 5 weeks · 52 mentor hours
- Schedule
- Mixed
- Level
- Intermediate
- Price
- 13,800 THB (informational)
Hibernate 6 · Postgres · Testcontainers
Description
Using a domain modeled on supply-chain manifests, you chase N+1 queries, rewrite fetch plans, and compare Hibernate statistics against SQL explain plans captured from Postgres 16.
What arrives in your repo
- Entity graph design workshop
- Statement batching with realistic payloads
- Second-level cache eviction drills
- Flyway migration ordering clinic
- Pessimistic vs optimistic locking matrix
- Micrometer metrics for pool saturation
- Mentor review on transaction boundaries
Outcomes we expect you to evidence
- Document fetch strategy per aggregate root
- Produce a before/after latency spreadsheet
- Recommend pool sizing with measured evidence
Responsible instructor
Lada Sooksan
Data layer consultant for wholesale distributors.
FAQ
We ship Docker Compose with seeded data; you may also point to a read-only clone.
Concepts transfer, but SQL labs target Postgres dialect.
We skip sharding strategies — only single-region topology.
Experience notes
“JPA + Hibernate Tuning batching lab exposed a silent cart merge bug — worth the late mentor replies.”
“Fetch graph section dense; appreciated the printable cheat sheet.”