When indexing documents in ElasticSearch we could put them into different indices and use aliases for searching. For example, we could index them by date and remove old data by removing the whole index. We'd also be able to switch indices without any downtime.
Do you see any other advantages when using multiple indices? Would it give us some performance improvements? Or maybe the opposite?
How about deleted documents? Does ElasticSearch free the used disk space on its own after removing them? If not, wouldn't it be better to have many smaller indices to be able to optimize them from time to time? I think optimizing one big index would take much time and the index will not be available for use at this time, right?
