Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

YouTube

GC Tuning Confessions of a Performance Engineer

Strange Loop Conference via YouTube

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore garbage collection tuning and performance engineering in this 43-minute conference talk from Strange Loop. Dive into performance metrics, tuning tradeoffs, and goals while comparing three garbage collectors in OpenJDK's HotSpot JVM: Parallel Old, CMS, and G1. Learn about performance planning, experiment design, and infrastructural needs for effective tuning. Discover insights on response time metrics, GC performance engineering, and the realities of ideal garbage collection. Examine the tradeoffs between throughput and latency, and understand the characteristics of different collectors. Investigate topics such as Java heap management, concurrent mode failures, regionalized heaps, and humongous objects. Gain practical knowledge on generation sizing, scaling, reference processing times, and GC overhead. Explore tuning parameters for the Throughput Collector and G1 Collector, and find resources for further reading on garbage collection optimization.

Syllabus

Intro
About Today's Talk
Performance Requirements
Response Time Metrics
Performance Analysis
GC Performance Engineering
Fun Facts!
Ideal GC?
The Reality!
The Tradeoff
The Throughput Maximizer
Mr. Latency Sensitive
The Serial Collector
The CMS Collector
Java Heap
Contiguous Old Generation
Concurrent Mode Failures
The Garbage First Collector - Regionalized Heap
Avoiding Evacuation Failures
G1 Heap Waste Percentage
The Garbage First Collector - Humongous Objects
Generation Sizing
Scaling: GC Times vs Sys Times
Reference Processing Times
GC Overhead vs Elapsed Time
Old Generation & Collections
GC Tuneables - The Throughput Collector
GC Tuneables - The G1 Collector
Further Reading

Taught by

Strange Loop Conference

Reviews

Start your review of GC Tuning Confessions of a Performance Engineer

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.