hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alok Singh <aloksi...@gmail.com>
Subject Re: ColumnSuffixFilter in HBase
Date Wed, 11 Feb 2015 18:44:20 GMT
A simple String suffix check will be definitely faster, but I doubt
you will notice the difference as most of the time will be spent
loading every column of every row to do the check.

Just for the fun of it, I did a quick comparison of regex "^.*xyz$"
with String.endsWith("xyz"): regex is about ~10x slower (2Million/sec
vs 20Million/sec).

Alok

On Tue, Feb 10, 2015 at 10:56 PM, anil gupta <anilgupta84@gmail.com> wrote:
> Completely missed out on QualifierFilter. Thanks Alok.
> Since QualifierFilter is going to use regex. I would say that
> QualifierFilter wont be super efficient. At present, i want to do this
> quickly.
> In near future, if time permits, i will try to write ColumnSuffixFilter and
> use it.
> An observation regarding naming, "ColumnPrefixFilter" and "QualifierFilter"
> names are not coherent. IMHO, we should either call ColumnQualifier as
> "Column" or "Qualifier". Using both interchangeably creates confusion.
>
> On Tue, Feb 10, 2015 at 9:57 PM, Jean-Marc Spaggiari <
> jean-marc@spaggiari.org> wrote:
>
>> Oh, you're right? I read the question too quickly and skipped the column
>> information... FuzzyRowFilter is only for the key.
>>
>> 2015-02-11 0:53 GMT-05:00 Alok Singh <aloksingh@gmail.com>:
>>
>> > Doesn't the FuzzyRowFilter only work on rowkeys? Not sure how it would
>> > apply to column qualifiers.
>> >
>> > Alok
>> >
>> > On Tue, Feb 10, 2015 at 9:48 PM, Jean-Marc Spaggiari
>> > <jean-marc@spaggiari.org> wrote:
>> > > Regex are more expensive that "basic" filters for sure. But it's a
>> simple
>> > > solution. You can implement a ColumnSuffixFilter. If you keys are
>> always
>> > > the same length, you can look at this:
>> > >
>> >
>> https://hbase.apache.org/0.94/apidocs/org/apache/hadoop/hbase/filter/FuzzyRowFilter.html
>> > >
>> > > JM
>> > >
>> > > 2015-02-11 0:44 GMT-05:00 hongbin ma <mahongbin@apache.org>:
>> > >
>> > >> Will there be any performance issues ? i'm curious if there's an
>> > efficient
>> > >> way of implement such kind of filer.
>> > >>
>> > >> On Wed, Feb 11, 2015 at 1:39 PM, Alok Singh <aloksingh@gmail.com>
>> > wrote:
>> > >>
>> > >> > You could use a QualifierFilter with a RegexStringComparator to
do
>> the
>> > >> > same.
>> > >> >
>> > >> > Alok
>> > >> >
>> > >> > On Tue, Feb 10, 2015 at 7:23 PM, anil gupta <anilgupta84@gmail.com>
>> > >> wrote:
>> > >> > > Hi,
>> > >> > >
>> > >> > > I want to get all the columns of a row that ends with "xyz".
I
>> know
>> > >> there
>> > >> > > is ColumnPrefixFilter. Is there any other column filter in
HBase
>> > that
>> > >> can
>> > >> > > help me simulate feature of suffix matching?
>> > >> > >
>> > >> > > In my opinion, ColumnSuffixFilter would be a good addition
to
>> > Filters
>> > >> of
>> > >> > > HBase.
>> > >> > >
>> > >> > > --
>> > >> > > Thanks & Regards,
>> > >> > > Anil Gupta
>> > >> >
>> > >>
>> >
>>
>
>
>
> --
> Thanks & Regards,
> Anil Gupta

Mime
View raw message