The original discussion is here.
We have a doc for partition support, but it only describes partition support for HadoopIndexTask while indexTask also supports it. Also it doesn't describe what users can expect with those partitioning in terms of performance benefit or disk space utilization.
I think it makes sense to add a new doc describing all missing stuffs about partition support and link it to HadoopIndexTask and indexTask pages.
The original discussion is here.
We have a doc for partition support, but it only describes partition support for HadoopIndexTask while indexTask also supports it. Also it doesn't describe what users can expect with those partitioning in terms of performance benefit or disk space utilization.
I think it makes sense to add a new doc describing all missing stuffs about partition support and link it to HadoopIndexTask and indexTask pages.