Articles by Cindy Sridharan
Logs and MetricsLogs and metrics are two distinct entities that are often confused or conflated. Observability is key to gaining visibility into modern day applications and infrastructure, and logs and metrics are two of the three pillars of observability. Logs are an immutable record of discrete events and are usually emitted in plaintext, structured, or binary formats. Logs are great for exploratory analysis of outliers and anomaly detection, but can be expensive to process and store. Metrics are numbers measured over intervals of time and are optimized for storage and enable longer retention of data. Metrics are better suited for monitoring and profiling purposes and are more malleable to mathematical and statistical transformations.
- Monitoring and Observability
Cindy contrasts the terms "monitoring" and "observability", where "monitoring" shows the overall health of systems and "observability" provides highly granular insights into the behavior of systems along with rich context, for debugging.
- Know how your org works (or how to become a more effective engineer)
Cindy explains how almost everything at a company "is political, and beyond a certain level, advancing further requires getting really good at playing this game." She then explains various aspects of navigating organizational politics: • The Mirage of Aspiration • Know How Your Org Works • Soft Skills Are Hard Skills • Understand Implicit Hierarchies • Cultures: Top-Down, Bottom-Up, and Both • Get comfortable with the “mess” • Look For Small (And Any) Wins • Understand Org Constraints and Manage Your Expectations
- Testing in Production, the safe way
Whether or not you choose to test in production based on reading this, this article will give you a fundamental understanding of real-world testing constraints. It covers: • Why test in production when one can test in staging? • The Art of Testing in Production • The Three Phases of “Production”: Deploy, Release, Post-Release • Testing in Production during the Deploy Phase