Hi Denis.

Thank you for your comments.

As you say, it would be better that Spark runs on container rather than VMs.
However, currently Meteos doesn't support container environment, because Sahara 
doesn't support.
Meteos depends on Sahara about environment which Spark runs.
I hope that Sahara support Container (Magnum) in future.

I have not evaluated the performance of Meteos in large-scale environment.
If performance problem occurs in VM environment, some kind of action will be 
needed ( for instance, participate in Sahara project to develop magnum or 
ironic driver).

Thanks,
Hiroyuki Eguchi

<[email protected]>
<[email protected]>


From: Denis Makogon [mailto:[email protected]] 
Sent: Saturday, December 17, 2016 6:10 AM
To: OpenStack Development Mailing List (not for usage questions)
Cc: Eguchi Hiroyuki(江口 裕之)
Subject: ##freemail## Re: [openstack-dev] [meteos] Meteos Released !!

Hello Hiroyuki.

Congrats on public release. So, i'd like to clarify few things. I know that for 
running ML jobs Meteos needs Spark instances and Meteos talks to Sahara to 
deploy them. So, i'm not quite familiar with infrastructure drives in Sahara, 
but would that make sense to work with Docker containers rather than deploying 
virtual machines, the thing is spinning up a VM takes a lot time, but Docker 
container takes 300ms to start.

Kind regards,
Denis Makogon


2016-12-16 7:14 GMT+02:00 Hiroyuki Eguchi <[email protected]>:
Hi all,
 
I'm pleased to announce the release of Meteos.
 
Meteos is Machine Learning as a Service (MLaaS) in Apache Spark.
 
Meteos allows users to analyze huge amount of data and predict a value by data 
mining and machine learning algorithms. 
Meteos create a workspace of Machine Learning via sahara spark plugin and 
manage some resources and jobs regarding Machine Learning. 
 
Everyone can participate in this project as a user, developer, reviewer in the 
same way as another OpenStack projects.
 
Please give it a try.
If you find any requests and comments, please feel free to feedback. 
 
See the following documents to find the relevant information:
 
[Wiki]
https://wiki.openstack.org/wiki/Meteos
 
[Installation Document]
https://wiki.openstack.org/wiki/Meteos/Devstack
 
[Examples]
Predict a Sales Figures by using LinearRegression Model
https://wiki.openstack.org/wiki/Meteos/ExampleLinear
 
Make a Decision to buy a stock by using DecisionTree Model
https://wiki.openstack.org/wiki/Meteos/ExampleDecisionTree
 
Recommend a Movie by using Recommendation Model
https://wiki.openstack.org/wiki/Meteos/ExampleRecommend
 
Search Synonyms by using Word2Vec Model
https://wiki.openstack.org/wiki/Meteos/ExampleWord2Vec
 
 
Thanks.
Hiroyuki Eguchi

__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: [email protected]?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: [email protected]?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Reply via email to