Hi, I have a MSSQL Server 2016 database table with 1.5TB of uncompressed text
varchar(max) with an integer ID field.
How much space would be required to copy the data from MSSQL Server to ElasticSearch (on Windows server)?
I am trying to create only one node to get started.
This is impossible to guess. It depends on so many factors.
The only way to know is to test it on a significant subset.
You mean, e.g.) extract about 10 samples of different 10Gs of texts into ES to see how they fan out?
What are the factors I should consider before I try those samples?
(I can think of changing an analyzer setting)
I mean extract something like 10% of the database to have something significant.
Analyzers, multi fields, _all field... all those have an impact. The content as well. Like if in a field you index only 5 different values or 500000 differentl values.
Thank you @dadoonet for giving me a guidance on how to get started.
I will try out different combinations to find good use cases for me.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.