@ankurm

Ankur in Hibernate

Master Hibernate 7 Connection Pooling with HikariCP: The Definitive Performance Guide

Are you struggling with sluggish database response times or "Connection is closed" exceptions in your Java logs? In modern enterprise applications, your database connection pool is the heart of your infrastructure. If that heart beats too slowly, your entire system suffers. In this guide, we explore how to integrate Hibernate 7 with HikariCP—the "zero-overhead" connection pool—to achieve maximum throughput and reliability. The Problem: The Latency of Connection Handshakes Database connections are heavy. When Hibernate needs to execute a query without a pool, it must perform a series of time-consuming steps: Open a network socket to the DB server. Complete a TCP/IP handshake. Negotiate SSL/TLS security. Authenticate the database user. Under high load, these milliseconds accumulate, leading to massive latency spikes. Without a pool, your application spends more time "connecting" than actually "querying." The Agitation: Why "Old School" Pools are Falling Behind For years, developers relied on pools like c3p0 or DBCP. While reliable, these libraries were built for an era of lower concurrency. They often suffer from: Internal Locking: Threads frequently block each other just to "borrow" a connection. Complexity: Dozens of confusing parameters that lead to misconfiguration. Size: Heavyweight codebases that increase your application's memory footprint. The Solution: Hibernate 7 + HikariCP HikariCP is widely recognized as the fastest connection pool available for the JVM. It is built on highly optimized, lock-free data structures and micro-benchmarked to ensure near-zero overhead.

Ankur in Hibernate

Mastering Hibernate 6 L2 Caching with Ehcache 3: The Definitive Guide

Hibernate 6 introduced a major architectural shift. By moving to the Jakarta EE namespace and embracing the JCache (JSR-107) standard, it changed how we interact with caching providers. If you are using Hibernate 6, the legacy hibernate-ehcache dependency is dead. To achieve high-performance data access today, you need the modern JCache bridge. The Problem: Database Bottlenecks in Hibernate 6 Even with the performance improvements in Hibernate 6’s new SQM (Semantic Query Model), database latency remains the primary bottleneck for scaling. Without a Second Level (L2) Cache, every time a new Session (EntityManager) is opened—even for the same user—Hibernate is forced to hit the database for data that hasn't changed. This results in redundant SQL SELECT statements, higher DB CPU usage, and increased costs in cloud environments where you pay for IOPS and database instance sizing. The Solution: Hibernate 6 + Ehcache 3 (JSR-107) The modern solution for Hibernate 6 is to use Ehcache 3 as a JCache provider. This allows Hibernate to offload entity and collection state to memory, sharing it across all sessions in the SessionFactory. Prerequisites Java 11+: Hibernate 6 requires a minimum of Java 11 (it is Hibernate 7 that raises the baseline to Java 17). Jakarta Persistence 3.x: The modern jakarta.persistence namespace. Step 1: Hibernate 6 Dependencies In Hibernate 6, you must use the hibernate-jcache module. Crucially, your Ehcache dependency must include the jakarta classifier to avoid namespace conflicts.

Ankur in Hibernate

Master Hibernate 7 Ehcache 3 Configuration: High-Performance Caching with Jakarta Persistence

Are you struggling with database performance in your modern Java applications? As systems scale, database bottlenecks remain the primary cause of latency. While Hibernate 7 introduces massive improvements in query generation and Jakarta Persistence compatibility, the secret to true high-concurrency performance lies in the Second-Level (L2) Cache. In this guide, we will configure Ehcache 3 — the industry standard for JVM-level caching — as the L2 cache provider for Hibernate 7, using the modern JCache (JSR-107) bridge.

Ankur in Hibernate

Hibernate 7 Second-Level Cache: When to Turn It On, How to Configure Ehcache 3, and the Three Ways It Will Stale Your Data

The question that matters before enabling L2 cache is not "can I make it faster" but "can I tolerate stale data, and for how long". The cache sits between Hibernate and the database, serving data that may have been written by a different JVM process, a batch job, a DBA running a script, or another application instance. Every one of those paths can invalidate the cache without Hibernate knowing. If your answer to the staleness question is "no, I cannot tolerate any staleness" — skip L2 entirely. If the answer is "yes, but only for these entities and within these bounds" — read on. This post covers what L2 actually stores, how to configure Ehcache 3 as a JCache provider, what changed between Hibernate 6 and 7, the three specific staleness modes that catch production teams off guard, and why the query cache is almost always the wrong additional layer.

Ankur in Hibernate

The Hibernate First-Level Cache Explained (It’s Not What You Think It Is)

Most developers who have used Hibernate for any length of time know the first-level cache exists. Ask them to describe it and you'll hear something like: "It's a cache Hibernate uses so it doesn't hit the database twice for the same row." That's technically correct, but it misses almost everything that matters. The first-level cache is not a feature you enable, a setting you tune, or an optional layer you bolt on for performance. It is the persistence context itself — always-on, transaction-scoped by default in Spring, and the thing that makes dirty checking, identity guarantees, and cascade operations possible. If you have ever used em.find(), you have used it. If you have ever hit an OutOfMemoryError in a batch job that loaded 200,000 entities, the first-level cache is why. This post is a deep look at the mechanics: what the persistence context stores, when it is consulted, how EntityKey works, and the failure modes that catch experienced developers off guard.

Ankur in Hibernate

Stored Procedures with Hibernate 7: @NamedStoredProcedureQuery, StoredProcedureQuery, and the Traps That Bite

Merges this site's two stored-procedure articles into one, and for the first time actually runs the examples -- against HSQLDB 2.7.3 with real SQL/PSM IN/OUT/INOUT procedures instead of an unexecuted MySQL listing. Covers @NamedStoredProcedureQuery, the programmatic StoredProcedureQuery API, and Hibernate-native ProcedureCall, plus the two findings that matter most: a wrong parameter name binds positionally instead of failing, and no stored procedure call auto-flushes pending changes -- not even with addSynchronizedEntityClass declared.

Ankur in Hibernate

The Ultimate Guide to Hibernate 7 Criteria Queries: Master Dynamic, Type-Safe Persistence

Writing dynamic queries with string-based HQL is fragile. One typo, one missing space, and your application fails at runtime. In enterprise applications—where filters change based on user input—this often turns into a mess of concatenated strings that are hard to test and harder to maintain. Hibernate 7’s Criteria API solves this by letting you build queries programmatically using a type-safe, refactoring-friendly API. With alignment to Jakarta Persistence 3.2+ and Hibernate 7’s improved Semantic Query Model (SQM), Criteria queries are now more predictable and production-ready than ever. In this guide, we’ll walk through Criteria Queries step by step—from basic selection to joins, analytics, and bulk operations—so you can confidently use them for real-world, dynamic data access. Why the Criteria API Exists In a typical enterprise application, search filters are rarely static. Users toggle checkboxes, select ranges, and combine conditions. Expressing this logic with raw HQL usually results in brittle string concatenation, runtime-only failures, and code that becomes impossible to refactor safely. The Criteria API replaces query strings with a structured, object-based model. Instead of writing text, you assemble query parts using Java objects. This shifts many errors from runtime to compile time and makes your persistence layer safer and easier to evolve.

Ankur in Hibernate

The Ultimate Guide to Hibernate Query Language (HQL) in Hibernate 7

Hibernate Query Language (HQL) lets you query your domain model instead of raw database tables, eliminating brittle SQL strings and manual result mapping. With the Semantic Query Model (SQM) engine (introduced in Hibernate 6, refined in 7), HQL is now more type-safe, more predictable, and better optimized than ever. In this guide, you'll learn how to use HQL effectively — covering pagination, joins, aggregates, and the most common pitfalls developers face moving from SQL to HQL.

Ankur in Hibernate

BLOB and CLOB in Hibernate 7: Streaming vs Eager, and the OOM You Didn’t See Coming

A list endpoint returned a page of 100 products. Simple enough — a cheap SELECT should be fast. But the page timed out, the heap spiked to 4 GB, and the GC ran continuously. The cause: the Product entity had an @Lob byte[] thumbnail field mapped with default eager fetching. Each of the 100 products loaded its thumbnail — averaging 40 MB each — all at once, into the heap. 100 rows × 40 MB = 4 GB from a list query that didn't display thumbnails. This is the OOM you don't see coming because the entity mapping looks harmless. This post covers the difference between byte[] (always eager), Blob with bytecode enhancement (genuinely lazy), and streaming (no heap allocation at all) — with approximate memory numbers for each. If you are building modern Java applications, handling BLOB and CLOB with Hibernate 7 is a skill you cannot ignore. In this guide, we will dive deep into how to efficiently map, persist, and retrieve binary and character data using the latest Hibernate standards aligned with Jakarta Persistence 3.2+. The Problem: The "Out of Memory" Nightmare Storing small strings like usernames or emails is easy. But what happens when your data grows to megabytes? Traditional mapping techniques often try to load the entire object into the JVM's memory. Imagine a scenario where 100 concurrent users try to download a 50MB PDF. If your application is configured to load the entire file into a byte[], your server will attempt to allocate 5GB of RAM instantly. In most environments, this leads to the dreaded: java.lang.OutOfMemoryError: Java heap space This crashes your service and disrupts all users.

Ankur in Hibernate

Hibernate 7 Named Queries: Startup Validation Proven, and Retiring the “Pre-Parsed = Faster” Claim

A companion repository proves named-query startup validation with a real broken query (NamedQueryValidationException at boot vs. IllegalArgumentException/UnknownPathException at call time), measures cacheable=true actually populating the second-level cache, and retires the "named queries are faster because they're pre-parsed" claim after measuring a 1.06-1.09 ratio against identical inline JPQL -- noise, not a real effect.