Is there any documentation or examples that show how the graph builder can
be used in such a manner? I am not familiar with graph builder but would
like to use it for such purposes.
On Mon, Dec 30, 2013 at 11:19 PM, Datta, Kushal <kushal.datta(a)intel.com>wrote:
Thanks for your interest in Graph Builder 2.0.
Saving a graph in Titan is one of the features of this release, but not
the “only thing”.
Graph construction is a data parallel problem and our goal is to provide a
library of methods that one can easily use to create large scale graphs
using mapreduce. Pig provides a simple and efficient way to ingest data
from many different sources (such as databases, HBase, HDFS and local file
systems). Also, Pig is popularly used for data ETL. However there was no
libraries in Pig that can build a large scale graph. Hence we provided the
new Graph Builder library in Pig which provides you the following methods:
- Create property graph elements from Pig schema or HBase table
- Dedpulicate property graph elements (i.e., vertices and edges)
- Export property graphs as edge lists (in HDFS) or RDF (in HDFS)
- Export property graph in Titan (w/ HBase currently) which is the
best known graph database that can manage large scale graphs in Hadoop/HBase
In days to come we will keep on adding more features and functionalities
to this library such that it becomes more usable, easier to extend and
achieves high scalability.
What kind of functionality are you looking for?
*From:* GraphBuilder [mailto:firstname.lastname@example.org] *On
Behalf Of *jamal sasha
*Sent:* Saturday, December 28, 2013 6:09 PM
*Subject:* [GraphBuilder] Question on graphbuilder
Seems like only thing Graphbuilder does is save the data to titan from
pig and that too if the backend db is hbase?
Or am I missing something?
GraphBuilder mailing list
Zaloni, Inc. | www.zaloni.com
633 Davis Dr., Suite 200
Durham, NC 27713
p: 919.602.4965 x7020