predictionio-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Pat Ferrel <...@occamsmachete.com>
Subject Re: Can I train and deploy on different machine
Date Thu, 30 Mar 2017 18:52:13 GMT
In the thread below I answered this. 

" Note that the PredictionServer should be configured to know how to connect to Elasticsearch
and HBase and optionally HDFS, only Spark needs to be local. Note also that no config in pio-env.sh
needs to change, Spark local setup is done in the Spark conf, it has nothing to do with PIO
setup.””



On Mar 30, 2017, at 11:14 AM, Marius Rabenarivo <mariusrabenarivo@gmail.com> wrote:

For the host where we run the training, do we have to put the path to ES_CONF_DIR and HADOOP_CONF_DIR
in pio-env.sh even if we use remote ES and Hadoop clusters?

2017-03-30 22:09 GMT+04:00 Marius Rabenarivo <mariusrabenarivo@gmail.com <mailto:mariusrabenarivo@gmail.com>>:
Replace Haddop by Hadoop in the previous mail

2017-03-30 22:08 GMT+04:00 Marius Rabenarivo <mariusrabenarivo@gmail.com <mailto:mariusrabenarivo@gmail.com>>:
For the host where we run the training, do we have to put the path to ES_CONF_DIR and HADOOP_CONF_DIR
in pio-env.sh even if we use remote ES and Haddop clulsters?

2017-03-30 21:58 GMT+04:00 Pat Ferrel <pat@occamsmachete.com <mailto:pat@occamsmachete.com>>:
To run locally in the same process as pio delete those files and do not launch Spark as a
daemon, only use PIO commands.

We do not “re-deploy” we hot-swap the model that predictions are made from so the existing
deployment works with the new data automatically and without any down-time.

Re-deploying means stopping the deployed process and restarting it. This is never necessary
with the UR unless engine.json config is changed.


On Mar 30, 2017, at 12:47 AM, Bruno LEBON <b.lebon@redfakir.fr <mailto:b.lebon@redfakir.fr>>
wrote:

"Spark local setup is done in the Spark conf, it has nothing to do with PIO setup.  "

Hi Pat,

So when you say the above, which files do you refer to? the "masters" and "slaves" files ?
So I should put localhost in those files instead of the dns names I configured in /etc/hosts?
Once this is done, I'll be able to launch 
"nohup pio deploy --ip 0.0.0.0 --port 8001 --event-server-port 7070 --feedback --accesskey
4o4Te0AzGMYsc1m0nCgaGckl0vLHfQfYIALPleFKDXoQxKpUji2RF3LlpDc7rsVd -- --driver-memory 1G >
/dev/null 2>&1 &"
with my Spark cluster off ?

Also, I have the feeling that once the train is done, the new model is automatically deployed,
is that so? In the template Ecommerce recommendation ,the log was explicitly telling that
the model was being deployed, whereas in Universal Recommender the log doesnt mention an eventual
automatic deploy right after the train is done.

 


2017-03-29 21:25 GMT+02:00 Pat Ferrel <pat@occamsmachete.com <mailto:pat@occamsmachete.com>>:
The Machine running the PredictionSever should not be configured to connect to the Spark Cluster.

This is why I explained that we use a machine for training that is a Spark cluster “driver”
machine. The driver machine connects to the Spark cluster but the PredictionServer should
not. 

The PredictionServer should have default config that does not know how to connect to the Spark
cluster. In this case it will default to running spark-submit to launch with MASTER=local,
which puts Spark in the same process with the PredictionServer and you will not get the cluster
error. Note that the PredictionServer should be configured to know how to connect to Elasticsearch
and HBase and optionally HDFS, only Spark needs to be local. Note also that no config in pio-env.sh
needs to change, Spark local setup is done in the Spark conf, it has nothing to do with PIO
setup.  

After running `pio build` and `pio train` copy the UR directory to *the same location* on
the PredictionServer. Then, with Spark setup to be local, on the PredictionServer machine
run `pio deploy` From then on if you do not change `engine.json` you will have newly trained
models hot-swapped into all PredictionServers running the UR.


On Mar 29, 2017, at 11:57 AM, Marius Rabenarivo <mariusrabenarivo@gmail.com <mailto:mariusrabenarivo@gmail.com>>
wrote:

Let me be more explicit.

What I want to do is not using the host where PredictionServer will run as a slave in the
Spark cluser.

When I do this I got "Initial job has not accepted any resources" error message.

2017-03-29 22:18 GMT+04:00 Pat Ferrel <pat@occamsmachete.com <mailto:pat@occamsmachete.com>>:
yes

My answer below was needlessly verbose.


On Mar 28, 2017, at 8:41 AM, Marius Rabenarivo <mariusrabenarivo@gmail.com <mailto:mariusrabenarivo@gmail.com>>
wrote:

But I want to run the driver outside the server where I'll run the PredictionServer.

As Spark will be used only for launching there.

Is it possible to run the driver outside the host where I'll deploy the engine? I mean for
deploying

I'm reading documentation about Spark right now for having insight on how I can do it but
I want to know if someone has tried to do something similar.

2017-03-28 19:34 GMT+04:00 Pat Ferrel <pat@occamsmachete.com <mailto:pat@occamsmachete.com>>:
Spark must be installed locally (so spark-submit will work) but Spark is only used to launch
the PredictionServer. No job is run on Spark for the UR during query serving.

We typically train on a Spark driver machine that is like part of the Spark cluster and deploy
on a server separate from the Spark cluster. This is so that the cluster can be stopped when
not training and no AWS charges are incurred. 

So yes you can and often there are good reasons to do so.

See the Spark overview here: http://actionml.com/docs/intro_to_spark <http://actionml.com/docs/intro_to_spark>


On Mar 27, 2017, at 11:48 PM, Marius Rabenarivo <mariusrabenarivo@gmail.com <mailto:mariusrabenarivo@gmail.com>>
wrote:

Hello,

For the pio train command, I understand that I can use another machine with PIO, Spark Driver,
Master and Worker.

But, is it possible to deploy in a machine without Spark locally installed as it is use spark-submit
during deployment
and 
org.apache.predictionio.workflow.CreateServer
references sparkContext.

I'm using UR v0.4.2 and PredictionIO 0.10.0

Regards,

Marius

P.S. I also posted in the ActionML Google group forum : https://groups.google.com/forum/#!topic/actionml-user/9yNQgVIODvI
<https://groups.google.com/forum/#!topic/actionml-user/9yNQgVIODvI>







-- 
You received this message because you are subscribed to a topic in the Google Groups "actionml-user"
group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/actionml-user/9yNQgVIODvI/unsubscribe
<https://groups.google.com/d/topic/actionml-user/9yNQgVIODvI/unsubscribe>.
To unsubscribe from this group and all its topics, send an email to actionml-user+unsubscribe@googlegroups.com
<mailto:actionml-user+unsubscribe@googlegroups.com>.
To post to this group, send email to actionml-user@googlegroups.com <mailto:actionml-user@googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/actionml-user/504FC93C-1A95-4673-BCB2-1C1E8CA0D487%40occamsmachete.com
<https://groups.google.com/d/msgid/actionml-user/504FC93C-1A95-4673-BCB2-1C1E8CA0D487%40occamsmachete.com?utm_medium=email&utm_source=footer>.

For more options, visit https://groups.google.com/d/optout <https://groups.google.com/d/optout>.




-- 
You received this message because you are subscribed to the Google Groups "actionml-user"
group.
To unsubscribe from this group and stop receiving emails from it, send an email to actionml-user+unsubscribe@googlegroups.com
<mailto:actionml-user+unsubscribe@googlegroups.com>.
To post to this group, send email to actionml-user@googlegroups.com <mailto:actionml-user@googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/actionml-user/CAC-ATVEfJLZ0OE7fJy9dU-xDGqSNJGvA6VB6xND5vCo3rWFALg%40mail.gmail.com
<https://groups.google.com/d/msgid/actionml-user/CAC-ATVEfJLZ0OE7fJy9dU-xDGqSNJGvA6VB6xND5vCo3rWFALg%40mail.gmail.com?utm_medium=email&utm_source=footer>.
For more options, visit https://groups.google.com/d/optout <https://groups.google.com/d/optout>.


Mime
View raw message