Talks
Speakers
Events
Topics
Sign in
Home
Talks
Speakers
Events
Topics
Leaderboard
Use
Analytics
Sign in
Suggest modification to this talk
Title
Description
Your app is your business, so keeping it healthy is important. Unfortunately, most of the tools available today are more like your doctor verifying the fact that you've had a heart attackâafter it's happened. You can do better. In this session, you'll learn how to use metrics to be more proactive about monitoring your applications health, and to suss out the subtle but important warning signs that can help you prioritize developer time and improve the developer experience. We'll talk about how to instrument your code, what to measure, how to interpret the data, as well as how you can use the data to streamline your development process. Help us caption & translate this video! http://amara.org/v/FGb5/
Date
Summarized using AI?
If this talk's summary was generated by AI, please check this box. A "Summarized using AI" badge will be displayed in the summary tab to indicate that the summary was generated using AI.
Show "Summarized using AI" badge on summary page
Summary
Markdown supported
In this session at Rails Conf 2013, Yehuda Katz and Carl Lerche discuss the importance of proactive application monitoring to maintain app health. They emphasize that most traditional tools merely react to problems after they occur, like a doctor diagnosing a heart attack post-factum. Instead, developers are encouraged to gather meaningful metrics to detect issues before they impact user experience. Key points from the session include: - **Collect Data**: Emphasizing the need to measure anything that provides value, enabling quick validation of hypotheses about app performance. - **Traffic Monitoring**: Discussing the importance of tracking key business metrics such as revenue per minute to identify anomalies immediately after deploying changes. - **Response Times**: An emphasis on understanding average response times versus median and standard deviation to make informed decisions on performance. - **Long-Tail Distribution**: Presenting research showing that web response times do not adhere to a normal distribution; instead, they exhibit long-tail behavior where outliers are more common than expected. - **Real-World Implications**: They shared a case study demonstrating how measuring metrics like the number of users adding items to the cart can pinpoint where issues are arising in the checkout process. - **Performance Tools**: Introduction of a monitoring tool, Skylight, designed to visualize app performance data using histograms to illustrate how response times are distributed across a range of interactions. This allows for deeper insights into where optimizations are needed rather than focusing solely on averages. Ultimately, the presenters strongly conclude that measuring every aspect of application interaction is essential in fostering a healthy app environment. By adopting this proactive approach to monitoring and measuring, developers can significantly enhance user experience and improve the efficiency of debugging. Understanding and interpreting correct metrics ultimately promotes better software development practices.
Suggest modifications
Cancel
