APM/Logs (Ready) Flashcards
What is logging?
The collection of raw data between devices and software
What is a log?
Text record what happened and when. Come in all levels of detail and complexity to track processes
What generates logs?
Anything that uses technology
Why do we need logs?
Logs have traditionally been used to troubleshoot
Why tag logs?
Tags allow us to correlate the data across hosts, services, applications, etc
Why collect data from logs?
Data can be collected to trouble shoot, build dashboards, alerts, collect/track business metrics, and feed into machine learning to proactively identify patterns/anomalies
What is log ingestion?
The storing of logs in an archive
What is log indexing?
The creation of a searchable collection of certain logs, made readily available
Why ingest all logs?
Useful data collection and correlation, logs available as needed for reference, pattern/anomaly detection
Why index just some logs?
Not all logs are needed all the time
What happens to ingested logs?
Fed into live-stream, important data parsed, tagged
How do we store ingested logs?
In a Datadog hosted archive, AWS or GCP
What is log rehydration?
The ability to index archived logs as need
What is Logging Without Limits?
The separation of ingesting and indexing logs, rehydrating from archives as based on customer need
What is Application Performance Monitoring (APM)?
Actively monitoring the errors, latency, and traffic related to an application
Difference between an application trace vs span?
A trace is an application process/request. A span is a single step in a trace
An application trace shows us…
The time it takes to complete an application process/request
An application span shows us…
The time it takes to complete a single step in an application process/request
What do errors tell us with APM?
What processes are failing
What does latency tell us with APM?
How long processes are taking to complete
What does traffic tell us with APM?
How many requests the application gets
Why tag application spans?
Tags allow us to correlate the data across hosts, services, applications, etc
Why collect data from application spans?
Data can be collected to trouble shoot, build dashboards, alerts, collect/track business metrics, and feed into machine learning to proactively identify patterns/anomalies
What is Application Code Profiling?
Profiling collects representative samples of all your stack traces
What does Application Code Profiling show?
Which application functions (or lines of code) consume the most resources (Memory, CPU, etc)
What is stack tracing?
A time-order visual (flame graph) of the processes/functions associated with an application span
Why use code profiling?
Optimize the application code to improve overall application performance, minimize resource consumption
Why tag application code profiles?
Tags allow us to correlate data across hosts, services, applications, etc
