In addition to setting the inputformat and outputformat classes, there
seems to be an option to specify additional name,value pairs
in the jobconf ("-jobconf"). The input and output format classes need
additional arguments that are not set using the other streaming arguments.
One example is the list of columns to scan. Please have a look at the
"initJob" method in o.a.h.hbase.mapred.TableMap and
o.a.h.hbase.mapred.TableReduce as well as the junit test,
o.a.h.hbase.mapred.TestTableMapReduce for how these are used.
I have not used the streaming package so have not tried this myself.
Sent by: news firstname.lastname@example.org
Re: map/reduce with hbase
Please respond to
Basically what I am trying to do is access hbase from php sense I do not
know java and have not found it fun to learn
I was looking around and found thishttps://issues.apache.org/jira/browse/HADOOP-2171
but am unsure if it is what I thank it is looks like a way to access hbase
from a socket server I have coded socket server/client in the past in php
the question is how would I use this to get access from say php.
would I add this and rebuild the hbase file somehow?
Remember I am new to java so have vary little knowing on how to do stuff.
"Billy" <email@example.com> wrote in
I am using map/reduce with hadoop-0.15.0-streaming.jar to process the data
with php scripts. I have coded to process the data the blow is an example
of word counts from the input.
bin/hadoop jar contrib/hadoop-0.15.0-streaming.jar -mapper
/var/www/search/hadoop/wc-reducer.php -input input/* -output output
What I am looking to do is get and store the input and output from/in
Is that possible with hadoop-0.15.0-streaming.jar?
If so can I get a command line example?