Slow aggregations query


(Alexander Burakovskiy) #1

Hi,
I have a problem with queries with aggregations on a big index.
My aggregations are basically facets for students list, so I have high cardinality fields (education, skills, titles etc).
My environment:

CPU Count 4
System Memory 16Gb
EC2 Instance Type m4.xlarge

Count of documents 50M
Document has several nested objects, here is example
https://gist.github.com/aburakovskiy/a5088b87b79130d6be91ee9f981323c3

"education": [
{
"school": "Gemological Institute of America",
"degree": "Graduate jewller, Metal and Jewelry Arts, A",
"startDate": "2001-01-01T00:00:00+0000",
"endDate": "2002-01-01T00:00:00+0000"
},
{
"school": "University of Mumbai",
"degree": "Bachelors , Business/Commerce, General, A",
"startDate": "1998-01-01T00:00:00+0000",
"endDate": "2001-01-01T00:00:00+0000"
}
]

Here is example of query:
https://gist.github.com/aburakovskiy/b6ea9d7170e500ff9b66d6c6dff3dda7

And it took over 3s.

I've tried to enable "eager_global_ordinals", but it didn't increase performance.

Is it possible to improve queries performance (decrease query time to less then 0.5s)?


(system) #2

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.