Hi,
we have a lot of docs, and keeping all docs in ES would consume too much 
space, but we keep the data in text files, and would like to use ES for 
indexing and
kibana for analytics.
kibana needs docs with _source / raw body, right? especially the table 
widget?

since our workload is very heavy on writes and very light on reads, could 
we instead of _source just store a filename and byte offset pointer?
and have ES just read the document from a given file & byte offset when it 
needs them (assuming file is available on ES node).  or
i could write a network service that receives a filename & byte offset and 
returns the corresponding document from a file.

that way the raw doc wouldn't have to be in ES, but ES could still return 
the full docs at query time, and hand them off to kibana?
is that possible?

thanks,
Dieter

-- 
You received this message because you are subscribed to the Google Groups 
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/0c074659-42a8-41c1-8d36-f9003162b9cb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to