Avery,

> Communication between mappers is not part of the MapReduce computing model.  Therefore, it doesn't make sense for them to include it as it would unnecessarily complicate the fault-tolerance recovery.

I agree that it doesn't make sense to complicate things by introducing communication between mappers.

But, my original query was, why can't the RPC communication be avoided with mappers in Giraph similar to MR by not running multiple supersteps in a single map process. If I am not wrong Hadoop supports MMR (multiple Maps) type of jobs and each map can map to the computation phase in a single superstep. Agree, that there is an over head of launching maps again and again, but communication between the mappers can be avoided.

I was trying to figure out the rational behind the approach taken in Giraph.

Regards,
Praveen

On Sat, Dec 10, 2011 at 10:44 PM, Avery Ching <aching@apache.org> wrote:
On 12/9/11 10:22 PM, Praveen Sripati wrote:
Jack,

> Giraph maps do communicate: via RPC.  This is done repeatedly in every mapper, during the compute phase.  This is something that is not normal to MapReduce, it is special to Giraph.

There must have been definitely some thought around this. But, we can also have a mapper correspond to just the computation phase in a superstep and avoid communication between the mappers as in MapReduce. Later spawn another set of mappers for the next superset. There might be some reason why communication between mappers was avoided in MR.

Communication between mappers is not part of the MapReduce computing model.  Therefore, it doesn't make sense for them to include it as it would unnecessarily complicate the fault-tolerance recovery.


Any thoughts?

Regards,
Praveen

On Sat, Dec 10, 2011 at 10:35 AM, Jake Mannix <jake.mannix@gmail.com> wrote:


On Fri, Dec 9, 2011 at 8:16 PM, Praveen Sripati <praveensripati@gmail.com> wrote:
Jake,


> Let's not crosspost, please, it make the thread of conversation totally opaque as to who is talking about what.

Agree. I got it after the OP.


> There is only one set of map tasks for the Giraph job - those long-running map tasks run possibly many supersteps.

OK. But, map tasks don't communicate with each other. How are messages sent across in the communication phase of a super step that happens within a map?

Giraph maps do communicate: via RPC.  This is done repeatedly in every mapper, during the compute phase.  This is something that is not normal to MapReduce, it is special to Giraph.
 
> In Giraph, vertices can move around workers between supersteps.  A vertex will run on the worker that it is assigned to.

Is there any advantage of moving the processing of vertices from one worker to another. Can't there be affinity between a worker and the vertices it processes?

Often there will be affinity, but if the graph itself evolves during computation (some sort of iterative pruning or clustering), then moving around may make sense.  Also: if nodes die.
 
  -jake


Regards,
Praveen

On Fri, Dec 9, 2011 at 11:33 PM, Jake Mannix <jake.mannix@gmail.com> wrote:
[hama-user to bcc:]

Let's not crosspost, please, it make the thread of conversation totally opaque as to who is talking about what.

On Fri, Dec 9, 2011 at 1:42 AM, Praveen Sripati <praveensripati@gmail.com> wrote:
Thanks to Thomas and Avery for the response.

> For Giraph you are quite correct, all the stuff is submitted as a MR job. But a full map stage is not a superstep, the whole computation is a done in one mapping phase.

So a map task in MR corresponds to a computation phase in a superstep. Once the computation phase for a superstep is complete, the vertex output is stored using the defined OutputFormat, the message sent (may be) to another vertex and the map task is stopped. Once the barrier synchronization phase is complete, another set of map tasks are invoked for the vertices which have received a message.

In Giraph, each superstep does not lead to storage into an OutputFormat.  The data lives all in memory from the time the first superstep starts to the time the final superstep stops (except that for tolerance of failures, checkpoints are stored to disk at user-specified intervals).  There is only one set of map tasks for the Giraph job - those long-running map tasks run possibly many supersteps.
 
In a regular MR Job (not Giraph) the number of Map tasks equals to the number of InputSplits. But, in case of Giraph the total number of maps to be launched is usually more than the number of input vertices.

Number of maps > number of input vertices?  Not at all.  That would be insane.  We want to be able to run over multi-billion vertex graphs.  We're going to launch multiple billions of mappers?   The splitting of the data in Giraph is very similar to in a regular MR job, divide up your input data among the number of mappers you have, and you're off and running.
 

> Where are the incoming, outgoing messages and state stored
> Memory

What happens if a particular node is lost in case of Hama and Giraph? Are the messages not persisted somewhere to be fetched later.

If nodes are lost, the system has to back up to the most recent checkpoint, where graph state has been persisted to HDFS.  Messages are not currently persisted, but the state at which the graph was in to produce any messages was.
 
> In Giraph, vertices can move around workers between supersteps.  A vertex will run on the worker that it is assigned to.

Is data locality considered while moving vertices around workers in Giraph?

Data is all in memory, and typical graph algorithms are basically sending roughly the size of the entire graph (number of total edges) out over distributed RPC in any given superstep, so shuffling the graph around by RPC is not much more to do.
 

> As you can see, you could write a MapReduce Engine with BSP on top of Apache Hama.

It's being the done other way, BSP is implemented in Giraph using Hadoop.

I'll let the Hama people explain to you about how one would implement MR on top of Hama.  You are correct that in Giraph, the Hadoop JobTracker/TaskTracker and HDFS are used as substrate to help implement BSP (although I would not say that "MR" is being used to implement BSP, as there is no MR going on in Giraph).

  -jake
 


Praveen

On Fri, Dec 9, 2011 at 12:51 PM, Avery Ching <aching@apache.org> wrote:
Hi Praveen,

Answers inline.  Hope that helps!

Avery

On 12/8/11 10:16 PM, Praveen Sripati wrote:
Hi,

I know about MapReduce/Hadoop and trying to get myself around BSP/Hama-Giraph by comparing MR and BSP.

- Map Phase in MR is similar to Computation Phase in BSP. BSP allows for process to exchange data in the communication phase, but there is no communication between the mappers in the Map Phase. Though the data flows from Map tasks to Reducer tasks. Please correct me if I am wrong. Any other significant differences?

I suppose you can think of it that way.  I like to compare a BSP superstep to a MapReduce job since it's computation and communication.
- After going through the documentation for Hama and Giraph, noticed that they both use Hadoop as the underlying framework. In both Hama and Giraph an MR Job is submitted. Does each superstep in BSP correspond to a Job in MR? Where are the incoming, outgoing messages and state stored - HDFS or HBase or Local or pluggable?

My understanding of Hama is that they have their own BSP framework.  Giraph can be run on a Hadoop installation, it does not have its own computational framework.  A Giraph job is submitted to a Hadoop installation as a Map-only job.  Hama will have its own BSP lauching framework. 

In Giraph, the state is stored all in memory.  Graphs are loaded/stored through VertexInputFormat/VertexOutputFormat (very similar to Hadoop).  You could implement your own VertexInputFormat/VertexOutputFormat to use HDFS, HBase, etc. as your graph stable storage.

- If a Vertex is deactivated and again activated after receiving a message, does is run on the same node or a different node in the cluster?

In Giraph, vertices can move around workers between supersteps.  A vertex will run on the worker that it is assigned to.

Regards,
Praveen