Platform observability and SLAs in MLOps - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is platform observability?

Platform observability is the ability to understand the internal state of a system by collecting and analyzing data like logs, metrics, and traces to detect issues and improve performance.

beginner

Name three key data types used in observability.

The three key data types are logs (text records of events), metrics (numerical measurements over time), and traces (records of requests as they move through services).

beginner

What does SLA stand for and what is its purpose?

SLA stands for Service Level Agreement. It is a contract that defines the expected level of service, such as uptime or response time, between a service provider and users.

intermediate

How does observability help meet SLAs?

Observability helps meet SLAs by providing real-time insights into system health, enabling quick detection and resolution of problems to maintain agreed service levels.

beginner

Give an example of a metric that might be monitored to ensure SLA compliance.

An example is system uptime percentage, which measures how often the platform is available and running as expected.

Which of the following is NOT a primary data type used in platform observability?

A Backups

B Metrics

C Traces

D Logs

Backups are not a data type used for observability; logs, metrics, and traces are.

What does an SLA typically define?

A The hardware specifications

B The programming language used

C The number of developers on a team

D The expected service performance levels

An SLA defines the expected service performance levels like uptime and response time.

Why is observability important for platform reliability?

A It hides system problems

B It replaces testing

C It helps detect and fix issues quickly

D It increases system complexity

Observability helps detect and fix issues quickly, improving reliability.

Which tool would you use to collect metrics for observability?

A Prometheus

B Git

C Docker

D Jenkins

Prometheus is a popular tool for collecting and storing metrics.

What is a trace in observability?

A A type of log file

B A record of a request's path through services

C A backup of data

D A metric measuring CPU usage

A trace records the path and timing of a request as it moves through different services.

Explain how platform observability supports maintaining SLAs.

Think about how knowing system health helps keep promises to users.

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

Consider what each data type tells you about the system.