You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@accumulo.apache.org by "panqingcui@163.com" <pa...@163.com> on 2015/01/19 03:56:33 UTC

accumulo query

HI
   Queries in accumulo mode can only be based on the rowkey or a full table scan query? There is no other way? For example, I think according to query the value of a Family Qualifier?



panqingcui@163.com

Re: accumulo query

Posted by Josh Elser <jo...@gmail.com>.

Yes, the row is the only "index" over the data.

You can limit data from a table based on column family (or column family 
and qualifier), but it's typically less efficient than reads over the 
row. You can use the fetchColumnFamily or fetchColumn methods on 
(Batch)Scanner.

If you use locality groups[1], you can get more efficient reads over 
columns in a locality group.

[1] https://accumulo.apache.org/notable_features.html#locality-groups

panqingcui@163.com wrote:
> HI
>     Queries in accumulo mode can only be based on the rowkey or a full table scan query? There is no other way? For example, I think according to query the value of a Family Qualifier?
>
>
>
> panqingcui@163.com