Presentation: JIT vs AOT Performance With GraalVM
Share this on:
What You’ll Learn
- Hear about GraalVM, what is it good for from a performance point of view.
- Learn how to optimize your applications for better performance.
Abstract
In this session we are going to talk about various aspects of performance, such as peak throughput, startup, memory footprint and more, and how you can optimize your applications for them with GraalVM. GraalVM is a high-performance virtual machine, bringing new performance optimizations for individual languages and seamless interoperability for polyglot applications.
In particular, we’ll discuss JIT and AOT compilation and and talk about their advantages and trade-offs. Also, we’ll go through some practical examples comparing JIT compilation with HotSpot and GraalVM, JIT and AOT, improving AOT with profile-guided optimizations and more.
What is the work you're doing today?
I'm a developer advocate for GraalVM, and this means that I’m responsible for interacting with our developer community. So I help people to learn about GraalVM: what it is, what capabilities does it offer, how they can use GraalVM in various scenarios and make the most of it.
What are the goals you have for the talk?
I want to offer an additional point of view on performance. Performance is something people often talk about, but often focus on one or two metrics, and there might be more ways to look at it, and additional opportunities to optimize. We will go through some additional metrics, discuss how they are connected, and how you can optimize for each of them, specifically in the just-in-time compilation and ahead-of-time compilation modes.
Could you give me an example of, say, a new performance optimization that you can do in GraalVM that you can't do with the JVM C2?
GraalVM can offer the ability to compile Java applications ahead of time into a native binary. The reason why you might be interested in doing so is because you will get faster startup and will consume significantly less memory compared to traditional just in time (JIT) compilation setup. JIT is very good for optimizing for peak performance, but for startup and memory footprints the native images can offer very competitive advantages. This is something you might be specifically interested in if you are working with cloud and microservices, as these performance metrics may become more relevant for you.
You optimize for a smaller or a quicker startup time vs. the longer term performance that the classic JVM has. I got that right?
With GraalVM you can optimize for both. With JIT and AOT modes, you can choose whatever works for you and what your priorities for a particular application are. You can really do both, and we discuss how to do this, along with benefits and trade-offs of both ways.
What do you want people to leave the talk with?
The idea is to give the audience some food for thought and motivation to try some of the different ways to trade-off the performance dimensions when they go back to work, and gain performance improvements as the result.
Similar Talks
License Compliance for Your Container Supply Chain
Open Source Engineer @VMware
Nisha Kumar
Observability in the SSC: Seeing Into Your Build System
Engineer @honeycombio
Ben Hartshorne
Evolution of Edge @Netflix
Engineering Leader @Netflix
Vasily Vlasov
Mistakes and Discoveries While Cultivating Ownership
Engineering Manager @Netflix in Cloud Infrastructure
Aaron Blohowiak
Optimizing Yourself: Neurodiversity in Tech
Consultant @Microsoft
Elizabeth Schneider
Monitoring and Tracing @Netflix Streaming Data Infrastructure
Architect & Engineer in Real Time Data Infrastructure Team @Netflix
Allen Wang
Future of Data Engineering
Distinguished Engineer @WePay
Chris Riccomini
Mapping the Evolution of Socio-Technical Systems
Agile Methods Coach & Advocate for Woman in Tech
Cat Swetel
Coding without Complexity
CEO/Cofounder @darklang
Ellen Chisa
Tracks
Monday, 11 November
-
Architectures You've Always Wondered About
Hard-earned lessons from names you know on scalability, reliability, throughput, and performance.
-
Languages of Infrastructure
This track explores languages being used to code the infrastructure. Expect practices on toolkits and languages like Cloudformation, Terraform, Python, Go, Rust, Erlang.
-
Building & Scaling High-Performing Teams
To have a high-performing team, everybody on it has to feel and act like an owner. Organizational health and psychological safety are foundational underpinnings to support ownership.
-
Bare Knuckle Performance
Killing latency and getting the most out of your hardware
-
Ethics, Regulation, Risk, and Compliance
With so much uncertainty, how do you bulkhead your organization and technology choices? Learn strategies for dealing with uncertainty.
-
Software Supply Chain
Life of a software artifact from commit to deployment. Security, observability and provenance of the software supply chain.
Tuesday, 12 November
-
Microservices Patterns & Practices
What's the last mile for deploying your service? Learn techniques from the world's most innovative shops on managing and operating Microservices at scale.
-
Pushing the Web Forward: JavaScript, Frameworks, Transpilers, and WebAssembly
JavaScript is the language of the web. Latest practices for JavaScript development in and how transpilers are affecting the way we work. We’ll also look at the work being done with WebAssembly.
-
Practices of DevOps & Lean Thinking
Practical approaches using DevOps and a lean approach to delivering software.
-
Modern CS in the Real World
Thoughts pushing software forward, including consensus, CRDT's, formal methods & probabilistic programming.
-
Socially Conscious Software
Power of leadership, Engineering Metrics and strategies for shaping the org for velocity.
-
Modern Data Architectures
Today’s systems move huge volumes of data. Hear how places like LinkedIn, Facebook, Uber and more built their systems and learn from their mistakes.
Wednesday, 13 November
-
Production Readiness: Building Resilient Systems
Making systems resilient involves people and tech. Learn about strategies being used from chaos testing to distributed systems clustering.
-
JVM and its Ecosystem
JVM futures, JIT directions and improvements to the runtimes stack is the theme of this year’s JVM track.
-
Optimizing Yourself: Human Skills for Individuals
Better teams start with a better self. Learn practical skills for IC.
-
Living on the Edge: The World of Edge Compute From Device to Infrastructure Edge
IOT/IIOT, Federated Learning, Application Gateways, Serverless, Kubernetes, WASM/WASI all at the edge? This track explores what’s being done at the modern edge.
-
Trust, Safety & Security
Privacy, confidentiality, safety and security: learning from the frontlines.
-
Machine Learning for Developers
AI/ML is more approachable than ever. Discover how deep learning and ML is being used in practice. Topics include: TensorFlow, TPUs, Keras, PyTorch & more. No PhD required.