The “Indexed” Outcome Category

Data types which are subject to dynamic sampling are represented in two different outcome data categories: "total" and "indexed". You can think of the "total" category as counting items that are stored in aggregated form, in other words, as metrics. The "indexed" category counts items that are stored individually.

If an item is dropped before metrics extraction & dynamic sampling, it will show up as an outcome in both categories (for example, "transaction" and "transaction_indexed"). If an item is dropped by dynamic sampling, it will show up as a "filtered" outcome in the "indexed" category, but it lives on in the form of metrics, for which we emit an "accepted" outcome in the "total" category.

These charts illustrate the flow of data categories for transactions:

For a sampled transaction:

transaction

transaction

transaction_indexed

transaction

transaction_indexed

SDK

Relay

Dynamic Sampling

Store

transaction

transaction

transaction_indexed

transaction

transaction_indexed

SDK

Relay

Dynamic Sampling

Store

For a transaction filtered by dynamic sampling:

transaction

transaction

transaction_indexed

transaction

transaction_indexed

SDK

Relay

Dynamic Sampling

Store

Reject

transaction

transaction

transaction_indexed

transaction

transaction_indexed

SDK

Relay

Dynamic Sampling

Store

Reject

The following data categories have a corresponding "*_indexed" category:

  • transactions
  • spans
  • profiles
Help improve this content
Our documentation is open source and available on GitHub. Your contributions are welcome, whether fixing a typo (drat!) or suggesting an update ("yeah, this would be better").