Mastodon Feed
slightlyoff@toot.cafe ("Alex Russell") wrote:
The most important thing about a performance remediation effort isn't making the system faster, even though that's the most visible outcome.
Remediation is a chance to learn *why*, and then teach the team.
*Why* was this system performing the way it did? *Why* was it designed that way? *Why* wasn't there observability in place to catch this? *Why* will these changes make the system faster? *Why* will we avoid regression in the future?