Unify and manage your data

Proactive Monitoring System for Match Rules

Learn about the Proactive Monitoring System that manages match rules processing performance.

The Proactive Monitoring System continuously monitors match rules processing and automatically flags or bypasses rules that are likely to hinder performance. The system inspects match rules for potentially performance-hindering issues, such as an excessive number of match tokens after you change the data. It dynamically evaluates match rules against tenant-specific performance criteria. Depending on the inspection and its configuration, the system either bypasses a poorly performing rule until you take action or records the issue as performance-diminishing and notifies you without bypassing the rule.

Receive a notification whenever the Proactive Monitoring System bypasses or flags a poorly performing match rule by subscribing to match rule notifications.

The Match Rules tab in the Data Modeler highlights any bypassed match rules and indicates the performance issue that caused the Proactive Monitoring System to bypass it during match rules processing. All you need to do to get your match rule back in the game is fix the reported problem.

Let’s see how the Proactive Monitoring System evaluates match rules for potential performance issues. The first thing to know is the evaluation rules are not one-size-fits-all; they are specific to your tenant. Each time you load data into your system, for match rules that are processed over a period of time (identified as the Observation Window in this illustration), the Proactive Monitoring System runs inspections on records that were created, changed, or deleted:
Inspections on all match rules
  • Number of tokens generated for entities exceeds the set threshold:

    When a match rule generates large numbers of token phrases for many entities, the platform slows down due to an unusual number of match evaluations for entities.

  • Number of tokens shared between entities exceeds the set threshold:

    When token phrases are shared by a large number of entities, the platform must run an unusual number of match evaluations for entities, which causes excessive collisions.

  • Number of matches generated per entity exceeds the set threshold:

    When many entities receive an excessive number of matches from the same rule, the platform must run an unusual number of match evaluations, which can significantly slow down matching. This inspection does not apply to Automatically Group match rules, which are covered by the large groups inspection under Automatically Group match rules.

Inspections on Automatically Group match rules
  • Number of groups generated with many participants exceeds the set threshold:

    When an Automatically Group match rule generates many groups with 50 or more participants, it can significantly slow down the matching process.

If a match rule fails a token, matches-per-entity, or the Automatically Group inspection, the Proactive Monitoring System bypasses or flags the rule. The system highlights the rule on the Match Rules tab and generates a notification if you have subscribed to match rule notifications. After you resolve the reported problem by editing the match rule, the Proactive Monitoring System includes the match rule in match processing again.