How does it work with Solr Cloud?

[junminstorage]

Read https://github.com/SOLR4189/solcolator#possible-architecture, it mentioned for large number of queries, it suggests a secondary collection which I am not very clear about.

Questions:

1. Can you explain more about secondary collection or share me few pointers?
2. Will solcolator as a SOLR Update Processor affect or impact the performance of the main purpose of solr collection. e.g indexing the documents and supporting queries?
3. If we have couple millions of stored queries, can we index them in its own solr cloud and then run document against the cloud? and how to do it?

[SOLR4189]

Sorry, I didn't see your comment.

1. it depends on your expectations. If we speak about million queries it can impact the performance of the main collection, so in this case I advise to put Solcolator UP to different collection and set your Indexer to send documents to main Collection and collection with Solcolator.
2. Yes, if we speak about million queries. Also it's important number CPU you assign to your server. Luwak works good in parallel mode.
3. Yes, Solcolator collection can be multi-shards. For example, you can do that each shard reads all queries from shared folder (NFS) and then you run documents against this collection.

[shamili26M]

is reverse query possible in solcolator ?