3 articles by Kevin Guthrie.
Cloudflare's global network is immense but not limitless. As we look for small ways to trim our resource usage, we sometimes get lucky and we can cut significantly more. Here’s how we reduced one of our Pingora-based service's RAM usage with statistics.
Developers have a gut-felt understanding for performance, but that intuition breaks down when systems reach Cloudflare’s scale. We need the rigor of statisticians to understand latency’s impact on users. Here we discuss our work to communicate customer-centric measurements of latency in our network.
Pingora handles 35M+ requests per second, so saving a few microseconds per request can translate to thousands of dollars saved on computing costs. In this post, we share how we freed up over 500 CPU cores by optimizing one function and announce trie-hard, the open source crate that we created to do it.