Supercharge Your Kubernetes AI Deployments with Wasm
summary
Claims a small WebAssembly AOT-compiled inference of MobileNet runs 2–2.5× faster than the same Python container and shows the benchmark code. Demonstrates building a Spin/Slide Wasm module, using WASI for host calls, packaging with a Wasmtime-based image and running it on Kubernetes via Krustlet and containerd-wasm-shim (runtimeClass + dedicated Wasm node pool), plus deploying with the cave-wasm operator on Azure so Wasm and Linux containers run side-by-side.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:04What WebAssembly really is and its core features
- 3:56WebAssembly in cloud-native
- 6:24WebAssembly System Interface and its role
- 7:30WebAssembly safety
- 9:58WebAssembly System Interface for server-side use
- 10:24Benchmarking containers
- 12:02AI workload performance
- 12:44Demo setup
- 16:16Side-by-side run
- 17:20AKS demo walkthrough
- 21:41Deploying WASM workloads
- 24:13Runtime classes setup
- 26:25Kubernetes deployment YAML for WebAssembly
- 26:46KubeAwesome intro
- 27:29Running WASM jobs
- 28:11Summary
related talks
32mWebAssembly Based AI as a Service on the Edge with Kubernetes
33mWebAssembly based AI as a Service with Kubernetes
41mWebAssembly Based AI as a Service
33mLeveling Up Wasm Support in Kubernetes
33mKubernetes on the edge: running low-latency workloads with WebAssembly
01h 04mDevelop, Build & Deploy Lightweight AI Services with WebAssembly
26mPaving the Way of WebAssembly in the Intersection of Machine Learning and Cloud Native
31mBuilding Event-Driven WebAssembly on Kubernetes: Runtime, Observability, and…