Hi,
We are performing capacity planning in elastic to evaluate the storage sizes.
For this we indexed one doc with dummy fields (7 simple fields and 13 nested) took the size of the doc which is 1148 bytes.
Then we extrapolated this for 1000 docs which should be 1145Kb. But however, when we indexed 1000 docs (all fields populated for each doc) and made a call to cat api to get the size it gave 295Kb for pri.store.size.
Question is why are the extrapolated and actual sizes different and what else is included in the pri.store.size apart from docs?
Elasticsearch creates indices behind the scenes and uses compression, so storage is at low volumes not at all linear. Once you have indexed a good amount of data (maybe a few million records) I believe you will start seeing a more linear increase in size. This will depend on the data and often vary over time as segments periodically are merged to increase in size and reduce storage overhead.
2 Likes
@Christian_Dahlqvist thanks for the explanation.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.