Ready, Set, Go: WASM-Powered Containers Taking Your AI To New He...
summary
Demos compare a native Docker image (reported 1.21 GB vs 633 KB) and show WasmEdge containers cut payloads dramatically, reduce memory use, and convert cold starts into milliseconds (authors claim up to 100x faster startup) while trading some token-generation speed on raw CPUs. Runtime path uses WasmEdge + Wasm‑NN to route inference to GPU/TPU/FPGA through a host runtime, and Kraken provides P2P distribution so models are fetched once and shared across edge nodes instead of hitting a central registry.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:04AI deployment at edge
- 6:00WebAssembly momentum
- 8:28WebAssembly in cloud-native
- 11:43WasmEdge for AI workloads
- 14:10Hardware-accelerated AI
- 19:48Efficient model delivery
- 22:54Docker vs WasmEdge demo
- 30:04Q&A
related talks
31mCloud Native WASM: What's All the Hype?
38mCloud-Native WebAssembly: Containerization On the Edge
33mWebAssembly based AI as a Service with Kubernetes
41mOptimizing Microservices with WebAssembly and Linux Containers: Exploring Hybrid Contain...
32mWebAssembly Based AI as a Service on the Edge with Kubernetes
26mPaving the Way of WebAssembly in the Intersection of Machine Learning and Cloud Native
31mWith WasmEdge to New Shores
29mSupercharge Your Kubernetes AI Deployments with Wasm