I need to understand why creating 1 index for whole my data will be better than make the partition for multiple indices.
please look at that:
It was taken from the next link:
(part 1: https://www.elastic.co/blog/how-kenna-security-speeds-up-elasticsearch-indexing-at-scale-part-1)
I read about it in that blog in part 2
why is this ok for them and not for me
if I create one index and use routing then each father's children will be in the same shard and because of that in the same node, there will be no parallel search and I will get the same search results of 1 minute or worse..
please tell me why I'm wrong
if I have multiple indices, each index for one family, I could search on one index but in less amount of data and in parallel because Index will have more than one shard..