Mapping file and the disk space

[ivar (Ravi)]

I have an index which has around 10 fields that are marked as
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}

now I created another index with same mappings except that I marked all the text fields as just keyword
{
"type": "keyword"
}

now reindexed all the documents(500000 docs) from source to destination

and when I check the size of the index, I dont see any difference between new index and original index

I expected the new index should have less size as I removed 8 text fields ?

can someone explain this behaviour ?

[martinr_ubi (Martin R.)]

Hi,

You also removed the ignore_above: 256.
Meaning its now using the default of 2147483647.

What happens if you set the ignore_above : 256 in your new index mapping and rerun this experiment?

[ivar (Ravi)]

@martinr_ubi

ur new index mapping and rerun this experiment?

Hi Martin

I'm sure that all my Id fields will be < 256 characters
and text will not be split by standard analyzer I'm using