How Fast is FaaS? Reducing Cold Start Times in Knative
CNCF [Cloud Native Computing Foundation] via YouTube
Overview
Explore the challenges and solutions for reducing cold start times in Knative, a platform for building serverless-style systems. Dive into the intricacies of autoscaling, the cold start problem, and innovative approaches to minimize container startup latency. Learn about Knative's scaling mechanisms, image pools, language-specific optimizations, and advanced techniques like container freezing and warm memory. Discover how these strategies contribute to achieving a Function as a Service (FaaS) experience with faster response times. Gain insights into the KPERF tool and its role in optimizing Knative performance.
Syllabus
Intro
How does Knative scale to 1
What happens when we need that first spot
Image pools
Languages
Knobs
Istio
Container Freezer
Warm Memory
KPERF
KPerfect
Taught by
CNCF [Cloud Native Computing Foundation]