Re: Data modelling questions

Seconded. Deterministic materialized keys at specific time granularities are definitely the way to go. If your frequency is high enough you could r/w data at second or ms resolution directly into memory and then roll those up into higher time resolutions on disk. The value, as noted, could be js

Re: Data modelling questions

On 2/23/15 1:33 PM, Jason Campbell wrote: Thanks for the info. The model looks reasonable, but something I would worry about is the availability of the key data. For example, the timestamps and msg-ids should be known without key-listing Riak (which is always a very slow operation). There i

Re: YZ search schema question

Just for the query itself, then yep :). Zeeshan Lakhani programmer | software engineer at @basho | org. member/founder of @papers_we_love | twitter => @zeeshanlakhani > On Feb 24, 2015, at 10:11 AM, Michael Martin wrote: > > Hi Zeeshan, > > Actually, I was trying to make this harder than i

Re: YZ search schema question

Hi Zeeshan, Actually, I was trying to make this harder than it really is. The easiest way to do this appears to be simply ANDing the two values in the query parms. Thanks, Michael Martin On Tue, Feb 24, 2015 at 9:59 AM, Zeeshan Lakhani wrote: > Hello Michael, > > You should be able to then s

Re: YZ search schema question

Hello Michael, You should be able to then set a field to be multiValued in your schema that would capture your use-case. You can find more info here -> http://docs.basho.com/riak/latest/dev/advanced/search-schema/ and http://stack

YZ search schema question

Hi all, I have a need to index on two separate fields in a json document - rather, I need to concatenate the two and index on that. Short of duplicating both into a single new item in my json document, how would I go about doing that? Example: Given this JSON: { "parent": "/path/to/parent",