accumulo-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David Medinets <david.medin...@gmail.com>
Subject Re: row count
Date Thu, 18 Apr 2013 01:43:24 GMT
Could you layer a scan time SummingCombiner on top of the
FirstEntryInRowIterator?
I don't know how to actually do this, but instinct says it should work and
significantly reduce the traffic back to the client.


On Wed, Apr 17, 2013 at 10:42 AM, Keith Turner <keith@deenlo.com> wrote:

> On Tue, Apr 16, 2013 at 9:33 PM, Venkat <rkreddy@gmail.com> wrote:
> > I am sure this question has been asked several times but I could not get
> to
> > the answer using usual searches - which iterator is the right one to
> count
> > the number of rows for a given value or a pattern of value ?
>
> Take a look at org.apache.accumulo.core.iterators.FirstEntryInRowIterator.
>  Does anyone know why this is not in the user iterator package?  Is
> there an issue with it?  This will bring back the first key/value for
> each row, then you could count those on the client side.   This will
> work for a range.  For a pattern, David's suggestion of the regex
> filter may be useful.   You could also look in the
> org.apache.accumulo.core.iterators.user.RowFilter.
>
> You could use FirstEntryInRowIterator and RegEx or RowFilter, but you
> would have to be careful about the order of the iterators.
>
> >
> > Venkat.
>

Mime
View raw message