Optimization rules overview
This article covers platform-level optimization rules
Platform-level optimization rules filter events after they reach Sekoia, at the intake level. If you use the Sekoia Endpoint Agent, a separate set of agent-level optimization rules lets you filter events before transmission. See Reduce event volume with the Sekoia Endpoint Agent.
Intake optimization rules allow you to control the processing and handling of incoming events before they are subjected to detection. These rules can be tailored to specific needs using filters and actions to effectively manage event data.
This article covers platform-level optimization rules
Platform-level rules filter events after they reach Sekoia, at the intake level. The Sekoia Endpoint Agent has its own agent-level optimization rules that filter before transmission. See Reduce event volume with the Endpoint Agent.
When to use platform-level optimization rules
Use platform-level rules when you cannot filter at the source or forwarder level, or when your source is not the Sekoia Endpoint Agent. If you are unsure which approach fits your situation, start with Log volume reduction strategies.
Logic and execution
Optimization rules act as a pre-processing layer. When an event enters the platform, the system evaluates it against your defined rules before any detection logic or long-term storage occurs.
- Execution order: Rules do not follow a specific sequential execution order, such as priority or creation date.
- Quotas: There is no limit on the number of rules you can create per community or per intake.
- Rule types: Rules can be Global (at the workspace level) or Community-based (managed by individual communities).
Rule components
A rule is defined by the following components:
- Scope: You can target specific intakes using a Community UUID, Dialect UUID, or Intake UUID.
- Filters: A rule can accept several filters. All defined filters of a rule must match an existing field to apply.
Enriched fields are not available in filters
Rules are evaluated at parsing time, before enrichment runs. Fields added during enrichment, such as sekoiaio.tags.*, are not available when rules are evaluated and will never match. Use only fields present in the raw parsed event.
- Action: This specifies the particular operation executed to optimize the event.
Benefits
- Cost management: By ignoring high-volume, low-value events, you reduce the volume of data stored.
- Data consistency: You can delete non-standard fields to ensure logs align with official ECS or Sekoia Taxonomy.
- Processing efficiency: Actions like shrinking events or deleting message fields optimize the internal workflow and associated storage.
See also
- Investigate overusage to learn more about overusage and how to avoid it.
- Optimization rules technical reference to learn more about filter operators, action bitmask values and more.
- Create an optimization rule to start creating your rule thanks to our step by step guide.
- Log volume reduction strategies to align your rules on your usage.