Java & the JVM · topic 5 of 6

JVM

Memory areas, garbage collection, JIT, class loading, tuning.

Questions below are grouped from foundations to advanced. Work top to bottom if the topic is new to you, or jump to a level if you are brushing up.

Questions: 26
Difficulty mix: 4 easy · 13 medium · 9 hard

Topics covered

Troubleshooting7
Memory Areas5
Garbage Collection5
JIT4
GC Tuning3
Class Loading1
Runtime1

Foundations

Easy · 4 questions

Definitions and building blocks — start here if the topic is new to you.

JVM vs JRE vs JDK — what's the difference, and which do you ship in a Docker image?
Memory AreasEasy
How does Java achieve 'write once, run anywhere' via bytecode?
Memory AreasEasy
What are the main JVM memory areas, and which are shared vs per-thread?
Memory AreasEasy
Stack vs Heap — where do primitives, references and objects actually live?
Memory AreasEasy

Core concepts

Medium · 13 questions

The questions most interviews spend their time on: how it works and when to use it.

Why is the heap split into young and old generations, and what's a minor vs major GC?
Garbage CollectionMedium
How does garbage collection decide what to free — and what are GC roots?
Garbage CollectionMedium
What replaced PermGen with Metaspace, and how can you still get a Metaspace OOM?
Memory AreasMedium
How do you size the heap (-Xms/-Xmx), and what's special about JVMs in containers?
GC TuningMedium
How do you choose a garbage collector — Serial, Parallel, G1, ZGC or Shenandoah?
GC TuningMedium
What is a stop-the-world pause, and how do you reduce GC pause times?
GC TuningMedium
What are the different OutOfMemoryError types, and what does each tell you?
TroubleshootingMedium
How do you enable and read GC logs to diagnose a memory problem?
TroubleshootingMedium
Strong, weak, soft and phantom references — when do you use each?
Garbage CollectionMedium
How does JIT compilation work (C1/C2, tiered), and why is JVM warm-up a thing?
JITMedium
How does Class-Data Sharing (CDS/AppCDS) cut JVM startup time, and why does it matter for containers and serverless?
TroubleshootingMedium
How can a Stream or lambda closure quietly leak memory in a long-lived cache?
TroubleshootingMedium
How do you use a JVM shutdown hook to drain in-flight requests before exit?
RuntimeMedium

Advanced

Hard · 9 questions

Internals, trade-offs and edge cases — useful for senior rounds and follow-ups.

How does G1 GC work internally — regions, mixed collections and humongous objects?
Garbage CollectionHard
How do ZGC and Shenandoah achieve sub-millisecond pauses on huge heaps?
Garbage CollectionHard
What is escape analysis, and how does it let the JVM allocate objects on the stack?
JITHard
Walk me through diagnosing long GC pauses / high GC CPU in production.
TroubleshootingHard
What is off-heap / direct memory, and how do you track and bound it?
TroubleshootingHard
How does the class loading delegation model work, and why do enterprise apps customize it?
Class LoadingHard
GraalVM Native Image vs a jlink'd JVM runtime — what do you actually give up for instant startup?
JITHard
Why does timing a loop with System.currentTimeMillis() lie to you, and how does JMH avoid those traps?
JITHard
A Java service is pegging CPU with no errors logged — how do you find out why?
TroubleshootingHard

Google AdSense slot

Topics covered

Foundations

JVM vs JRE vs JDK — what's the difference, and which do you ship in a Docker image?

How does Java achieve 'write once, run anywhere' via bytecode?

What are the main JVM memory areas, and which are shared vs per-thread?

Stack vs Heap — where do primitives, references and objects actually live?

Core concepts

Why is the heap split into young and old generations, and what's a minor vs major GC?

How does garbage collection decide what to free — and what are GC roots?

What replaced PermGen with Metaspace, and how can you still get a Metaspace OOM?

How do you size the heap (-Xms/-Xmx), and what's special about JVMs in containers?

How do you choose a garbage collector — Serial, Parallel, G1, ZGC or Shenandoah?

What is a stop-the-world pause, and how do you reduce GC pause times?

What are the different OutOfMemoryError types, and what does each tell you?

How do you enable and read GC logs to diagnose a memory problem?

Strong, weak, soft and phantom references — when do you use each?

How does JIT compilation work (C1/C2, tiered), and why is JVM warm-up a thing?

How does Class-Data Sharing (CDS/AppCDS) cut JVM startup time, and why does it matter for containers and serverless?

How can a Stream or lambda closure quietly leak memory in a long-lived cache?

How do you use a JVM shutdown hook to drain in-flight requests before exit?

Advanced

How does G1 GC work internally — regions, mixed collections and humongous objects?

How do ZGC and Shenandoah achieve sub-millisecond pauses on huge heaps?

What is escape analysis, and how does it let the JVM allocate objects on the stack?

Walk me through diagnosing long GC pauses / high GC CPU in production.

What is off-heap / direct memory, and how do you track and bound it?

How does the class loading delegation model work, and why do enterprise apps customize it?

GraalVM Native Image vs a jlink'd JVM runtime — what do you actually give up for instant startup?

Why does timing a loop with System.currentTimeMillis() lie to you, and how does JMH avoid those traps?

A Java service is pegging CPU with no errors logged — how do you find out why?