Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Plain View
Pig >> mail # user >> Effective way to cross two relations


+
KALLURI, RAJESH K 2013-04-19, 04:17
Copy link to this message
-
Re: Effective way to cross two relations
I posted on this very same topic a few weeks ago with no response. It is
still an unresolved issue for me, so if anyone had any ideas it would be
greatly appreciated.

Interestingly enough I ran into issues right around the same size that you
are dealing with (50k rows) so I am wondering if it is an issue with how
Pig handles things. I'd recommend tuning some of the parameters that I
mention in my post (below) as it may help you complete the job.

http://search-hadoop.com/m/kJghFzruCA1/nested+cross&subj=Moving+Cross+of+Large+Data+to+be+Nested
On Thu, Apr 18, 2013 at 9:17 PM, KALLURI, RAJESH K (AG/1000) <
[EMAIL PROTECTED]> wrote:

> I have a relation of about 50000 tuples that I want to join to itself
> either by using a cross operator or something similar. Then I would be
> doing pair wise computation of half the matrix (avoiding comparing to self
> and duplicate).
>
> I was wondering what the most effective way to do this,  below is some
> pseudo pig latin.
>
>
> -- About 50,000 - 70,000 entries
> a = LOAD 'part-r-00000.txt' USING PigStorage()
> AS (id:long,  x:int, y:int);
> -- Same as a , About 50,000 - 70,000 entries
> b = LOAD 'part-r-00000.txt' USING PigStorage()
> AS (id:long,  x:int, y:int);
>
> jnd = join a by id , b by id;
> -- filter comparisons to self and duplicates from the matrix
> -- end up with 50000 X (50000-1)/2 entries
> filter_self = filter jnd by a::id != b::id and a::id > b::id;
>
> raw = foreach filter_self generate a::id as id1, b::id as id2, TOBAG(a::x,
> b::y) as z;
> -- group pairs for comparison
> grpd = group raw by (id1, id2);
> -- calculate similarity between id1 and id2 based on a udf
> prjctd = foreach grpd generate flatten(group), UDF(raw.z);
>
> This e-mail message may contain privileged and/or confidential
> information, and is intended to be received only by persons entitled
> to receive such information. If you have received this e-mail in error,
> please notify the sender immediately. Please delete it and
> all attachments from any servers, hard drives or any other media. Other
> use of this e-mail by you is strictly prohibited.
>
> All e-mails and attachments sent and received are subject to monitoring,
> reading and archival by Monsanto, including its
> subsidiaries. The recipient of this e-mail is solely responsible for
> checking for the presence of "Viruses" or other "Malware".
> Monsanto, along with its subsidiaries, accepts no liability for any damage
> caused by any such code transmitted by or accompanying
> this e-mail or any attachment.
>
>
> The information contained in this email may be subject to the export
> control laws and regulations of the United States, potentially
> including but not limited to the Export Administration Regulations (EAR)
> and sanctions regulations issued by the U.S. Department of
> Treasury, Office of Foreign Asset Controls (OFAC).  As a recipient of this
> information you are obligated to comply with all
> applicable U.S. export laws and regulations.
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB