In this episode of Cloud Performance Atlas, +Colt McAnlis helps a friend gain clarity into why her user’s 2G connections are sending her App Engine Instance count sky high. Can the costs get back down to earth? Stay tuned to find out.

Even after a service has been restored, SREs still have a bit of work to do. In this video, Liz and Seth discuss the postmortem process that SREs follow. Blameless postmortems and retrospectives are key to learning from failures and preventing recurrence. You will learn about the importance of conducting a postmortem, strategies for conducting a blameless postmortem, and techniques for trending retrospectives across your entire organization to gain better insights to prevent service disruptions in the future.

Aja built an interactive disruption detector panel for attendees at the Google I/O Conference to intentionally cause errors to happen to the system. This demo highlights the amazing real time monitoring feature of Stackdriver as it tracks all incoming errors and make things easier for developers to pinpoint the issue. Watch the video to learn more.

In the previous video, Liz and Seth discussed how to make systems observable and how observability helps us diagnose failing systems, but didn't cover what to do when an incident grows beyond the ability of one person to do it all. In this video, you learn about the most important part of the incident management process – humans.

Stackdriver Transparent SLI Monitoring provides detailed API level metrics for developers and IT teams to quickly diagnose a problem and pinpoint where the error is located, whether it is with your stack or with the service provider.

In their previous video, Liz and Seth reduced actionable alerts by focusing on Service Level Objectives (SLOs), but how can we make our systems observable, instead of only being able to debug what we've thought to monitor in the past?

Observability remains a key challenge as customers embrace DevOps. Join Daniel "Spoons" Spoonhower, the CTO and Founder of Lightstep, a Google Cloud customer, and Yuri Grinshteyn, a Google Cloud Customer Engineer to learn about how Lightstep was built on Google Cloud to enable you to monitor what matters most and diagnose anomalies within seconds across web, mobile, monoliths and microservices.

Join Stephanie and Mark as they introduce basic logging and monitoring with Stackdriver with a quick demo.

For a complete serverless solution you need to be able to monitor your serverless apps, learn when they are not functioning correctly, and debug them when necessary. This session teaches you how to use Google Stackdriver to monitor and troubleshoot serverless apps.

Preview of the Stackdriver: Qwik Start lab - part of the GCP Essentials Quest on Qwiklabs. Come in with little, or no prior cloud knowledge and come out with practical experience that you can apply to any Google Cloud project.