Events to Metrics in Chronologics

Events to Metrics Overview

• Events to Metrics allows users to turn logs or spans into metrics, making it easy to hold onto data for a long time without incurring high costs.
• Metrics are very easy to retain and can be queried in a very performant way across long periods of time.
• This feature unlocks historical reporting capabilities without the need for expensive indexing.

Prerequisites

• A metrics bucket needs to be set up in order for Events to Metrics to work.
• The metrics bucket can be set up in the Data Flow section, inside the Archive section.

Benefits

• Events to Metrics enables huge cost savings while retaining key information over a long period of time.
• The power of Chronologics lies in the ability to convert logs and traces into metrics without indexing, resulting in significant cost savings.

Converting Logs to Metrics

• In the Log Explore view, users can filter logs by a specific field, such as HTTP response took Ms.
• A query can be run to capture the field, and then an event to metrics category can be defined to capture the field.
• The metric series will start generating from the particular field when the "Save" button is clicked.
• The resulting metrics can be visualized in a custom dashboard.

Converting Spans to Metrics

• Spans can be converted into metrics in a similar way to logs.
• Users can extract the duration field from a span and generate metrics from it.
• The resulting metrics can be visualized in a custom dashboard.

Creating an Event to Metrics Feature

• To create an event to metrics feature, users need to hover over Data Flow, go to Events to Metrics, and click on "New Event to Metric".
• The first step is to enter basic details, such as the name and description of the event to metric.
• Users can then choose between logs and spans, and enter a query to filter the data.
• Metric fields can be defined, and labels can be added to the metrics.

Metric Fields

• Up to 10 different metric fields can be defined.
• Metric fields can be used to generate different metric series, such as maximum, minimum, and average.
• Advanced options allow users to customize the metric fields, such as adding histograms or aggregations.

Labels

• Labels are essentially values that can be added to metrics.
• Labels can be used to query and slice the data.
• Users need to be aware of the implications of adding labels, as they can affect the number of permutations.

Permutations

• Permutations refer to the different combinations of unique values for labels.
• The number of permutations can quickly increase if multiple labels are added.
• A graph is available to show the number of permutations per day, helping users to understand the impact of adding labels.

Best Practices

• It is recommended to make specific queries to avoid slowdowns.
• Users should be aware of the implications of adding labels and permutations.
• It is important to think about how the data will be used and define labels accordingly.### Permutations
• Permutations can quickly increase exponentially, making it possible to reach the 10 million permutations per account limit
• It is essential to carefully consider labels with unique values and how they impact permutation metrics

Managing Permutations

• Use the Matrix permutations graph to understand the impact of labels on permutations
• Allocate a daily permutation budget (e.g., 1100, 1200, etc.) to avoid reaching the limit
• Consider the permutation budget when adding labels, and evaluate whether they are necessary or can be optimized

Event to Metrics Configuration

• Configure events to metrics, setting the query, scope, and severities for logs
• Set services and actions for given spans
• Configure different values in the metrics fields (limited to 10 metric fields per program to metrics)

Metric Permutations and Labels

• Metric permutations are crucial, and their impact should be monitored
• The number of metric permutations can be viewed in the dashboard
• Labels can significantly affect permutations, making it essential to balance the number of labels and high cardinality labels with the permutation budget

Benefits of Custom Dashboards

• Cost-effective: storing data in object storage in the cloud account
• Prerequisite: having the metrics archive set up
• No additional cost impact: generate data without incurring extra costs
• Provides powerful, high-performance historical perception of data