Monitoring and observability in Prompt Engineering / GenAI - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is monitoring in DevOps?

Monitoring is the process of continuously checking the health and performance of systems using tools that collect and display data like CPU usage, memory, and errors.

intermediate

Define observability and how it differs from monitoring.

Observability is the ability to understand the internal state of a system by analyzing its outputs like logs, metrics, and traces. Unlike monitoring, it helps find unknown problems, not just known issues.

beginner

Name the three main types of data used in observability.

The three main types are:<br>1. Metrics - numbers showing system performance<br>2. Logs - detailed event records<br>3. Traces - records of requests moving through services

beginner

Why is alerting important in monitoring?

Alerting notifies the team immediately when something goes wrong, so they can fix issues before users notice problems.

beginner

Give an example of a popular monitoring tool and one observability tool.

Monitoring tool example: Prometheus<br>Observability tool example: Jaeger (for tracing) or Grafana (for visualization)

Which data type in observability shows the path of a request through multiple services?

A Alerts

B Logs

C Metrics

D Traces

Traces record the journey of a request across different parts of a system.

What is the main goal of monitoring?

A To fix bugs in code

B To continuously check system health and performance

C To deploy new software versions

D To write documentation

Monitoring focuses on tracking system health and performance metrics.

Which of these is NOT a typical observability data type?

A Backups

B Metrics

C Traces

D Logs

Backups are not part of observability data; they are copies of data for recovery.

Alerting helps teams by:

A Automatically fixing issues

B Backing up data

C Notifying when problems occur

D Writing code

Alerting sends notifications so teams can respond quickly to issues.

Which tool is commonly used for monitoring metrics?

A Prometheus

B Jaeger

C Git

D Docker

Prometheus is a popular tool for collecting and querying metrics.

Explain in your own words the difference between monitoring and observability.

Think about how monitoring is like watching a dashboard, while observability is like investigating a mystery.

List and describe the three main types of data used in observability.

Imagine metrics as speedometers, logs as diaries, and traces as travel maps.