The page that you are currently viewing is for an old version of Stroom (7.1). The documentation for the latest version of Stroom (7.6) can be found using the version drop-down at the top of the screen or by clicking here.
Search Extraction
When indexing data it is possible to store (see Stored Fields all data in the index. This comes with a storage cost as the data is then held in two places; the event; and the index document.
Stroom has the capability of doing Search Extraction at query time. This involves combining the data stored in the index document with data extracted using a search extraction pipeline. Extracting data in this way is slower but reduces the data stored in the index, so it is a trade off between performance and storage space consumed.
Search Extraction relies on the StreamId and EventId being stored in the Index. Stroom can then used these two fields to locate the event in the stream store and process it with the search extraction pipeline.