0

I already tried cleo, not convinced majorly because of this bug that I recently encountered and yet have been unable to solve.

Also tried elasticsearch, but too complex to run even a single query and indexing and other features were pretty slow too.

So if anyone knows another better one or something I am missing out here in these two. Thanks.

Saurabh
  • 492
  • 3
  • 10

1 Answers1

0

Have you tried the Completion Suggester API? It is not complex and certainly not slow.

Dan Tuffery
  • 5,874
  • 29
  • 28
  • Hey Dan, yeah that was what I reverted to after Cleo, but being still new to the concept of shards and nodes. It seemed complex to me at the first sight. I saw on the [page](http://www.elasticsearch.org/blog/you-complete-me/) that to run even a single query on auto suggestion I need to first create a query mapping than manually insert all the name suggestions and then again run a big query to find the suggestions for a given name. May be I am missing something here and you can guide me in the right direction. Thanks. – Saurabh Jul 31 '14 at 17:25
  • For now forget about shards and nodes. Download Elasticsearch and start it with the defaults, run through the examples in the Completion Suggester page to see if it meets your requirements. The Completion Suggestion uses Finite State Transducers so it is very fast, it should not be slow, if you run into issues ask on here. If you decide the Completion Suggester meets your requirements then you can research nodes and shards – which give you the option to scale out if you need to, resulting in better performance for very large data sets. – Dan Tuffery Aug 01 '14 at 10:28