Hi Alessandro,
Thanks for replying.
I understand how Tool works. So now the thing is that I'm having trouble
understanding how the control flows within a superstep. The program starts
with the main class calling Run inside PageRankBenchmark. The options are
set and after that the VertexClass is set. I fail to understand when
exactly is the Compute function in PagerRankBenchmark is called. Besides
this, which all files are used in the execution of the PageRankBenchmark
example.
1. PageRankBenchmark.java ?
2. PageRankComputation.java ?
Thanks.
Ganesh.
On Wed, Aug 1, 2012 at 4:39 AM, Alessandro Presta <alessan...@fb.com> wrote:
> Hi Ganesh,
>
> The way our benchmarks are currently organized is the following: a
> Benchmark class extends Vertex (the main unit of computation in a Giraph
> job) and Tool (a Hadoop application that can be run on a cluster).
>
> Although the computation code is essentially the same,
> SimplePageRankVertex is only used as an example and test case, so it's not
> related to benchmarks.
>
> Alessandro
>
> From: Dodda Ganesh <doddagane...@gmail.com>
> Reply-To: "user@giraph.apache.org" <user@giraph.apache.org>
> Date: Wednesday, August 1, 2012 6:16 AM
> To: "user@giraph.apache.org" <user@giraph.apache.org>
> Subject: Help understanding Page Rank example.
>
> Hi Everyone,
>
> I am trying to understand the flow on control in giraph. I took the
> example of PageRank which was given with the code. However, I am really
> confused.
>
> if (!cmd.hasOption('c') ||
> (Integer.parseInt(cmd.getOptionValue('c')) == 1)) {
> job.setVertexClass(PageRankBenchmark.class);
> } else {
> job.setVertexClass(HashMapVertexPageRankBenchmark.class);
> }
>
> Here, the file 'SimplePageRankVertex' is not used at all. Can someone
> please help me ?
>
> Ganesh
>
