Hello, I have a question I have a big cluster wit...
# community-help
a
Hello, I have a question I have a big cluster with 16million data of books. Now each book has a description, the description is localized. Now some books have a locale like aa or cgg, now only 2 books have the aa locale and 10 the cgg, ... In the schema we have 20 extra fields due to these locales. Now not all books only a few have all the locale values. What does this mean for memory usage? Is there a cost to having a field in the schema that almost no book has? Or does this not take up any memory if a products does not have a field?
f
For your schema, those rare locale fields do not use RAM for every book just because they exist. If a book does not have shortDescription_aa or shortDescription_cgg, Typesense has nothing to index for that book in those fields. You only pay a small fixed cost for the field definition and empty index structures, plus memory for the few documents that actually contain values.