spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Grzegorz BiaƂek <>
Subject Spot instances on Amazon EMR
Date Thu, 18 Sep 2014 10:19:55 GMT
I would like to run Spark application on Amazon EMR. I have some questions
about that:
1. I have input data on other hdfs (not on Amazon). Can I send all input
data from that cluster to HDFS on Amazon EMR cluster (if it has enough
storage memory) or do I have send it to Amazon S3 storage and then load
this data on EMR cluster where I want to run my application?
2. Which nodes should be on-demand instances and which can be spot
instances (I don't want to spend to much money but I also lost my data or
have to recompute everything after spot instance interruption)?
3. Can I use Amazon S3 storage for input and output data to have less
on-demand instances and more spot instances? (Or maybe there is another
solution to lower costs)

I would like to run this application once and computation would take around
30h I think.

Could you answer on (at least some of) this questions?


View raw message