Range filter performance consideration

Elastic Stack Elasticsearch
1

Hi, I am currently make a lot of range filter in this way:

"filter":{
"range":{"date":{"gte":1438358400, "lte":1438358460}}
}

if we disable cache, does elasticsearch will scan every doc in my index to determine if it falls into the range in this filter?

2

That is correct.

3

Oh, that's really not a good news for me. Because I have daily indices of 250GB each and I also make a lot of range filter, most of the range are just less then 100 seconds. That means every docs in some index will be read again and again!
in sql database, I know I can setup some field be ordered and no need for full scan for query in my senario. Is there any way for elasticsearch to do in some way like this?