Effective transaction management with Spring
summary
Live Java demos use Spring Data JPA, breakpoints and JUnit to show isolation and propagation effects on concurrent transactions: a repeatable-read test blocks a concurrent update so a bid reads 160 twice while the updater waits until commit, then becomes 170; lowering to read‑committed allows intervening commits and non‑repeatable reads. A serializable test blocks table inserts; a propagation test with requires_new isolates log writes results in 3 items and 4 log rows after a duplicate failure. Lost updates are avoided by design.
AI-written summary · about
chapters (8)
- 0:00Introduction to talk
- 1:58ACID properties of transactions
- 8:25Transaction isolation levels explained
- 17:54Demo on isolation levels
- 24:47Spring transactional annotations
- 33:40Propagation in Spring transactions
- 36:53Demo on propagation
- 41:57Q&A and resources
related talks
33mEffective Transactions Management with Spring
01h 17mUsing @Transactional Like a Pro
34mThe Complexity of JPA
49mPerformance oriented Spring Data JPA & Hibernate
41mTransactions: myths, surprises and opportunities
41mConcurrency, Scalability and Transactions -- Myths and Surprises
58mMaintain Social Distance this Spring
44mPerformance oriented Spring Data JPA & Hibernate