amazon web services - How to run a Spark jar file from AWS Console without Spark-Shell -
i'm trying run spark application on aws emr console (amazon web services). scala script compiled in jar takes sparkconf settings parameters or strings:
val sparkconf = new sparkconf() .setappname("wikipediagraphxpagerank") .setmaster(args(1)) .set("spark.executor.memory","1g") .registerkryoclasses(array(classof[prvertex], classof[prmessage]))
however, don't know how pass master-url parameter , other parameters jar when it's uploaded , set-up cluster. clear, i'm aware if running spark-shell way, i'm windows user , current set-up , work i've done, useful have way pass master url emr cluster in 'steps'.
i don't want use spark-shell, have close deadline , have set-up way , feels small issue of passing master url parameter should possible, considering aws have guide running stand-alone spark applications on emr.
help appreciated!
here instructions on using spark-submit via emr step: https://github.com/awslabs/emr-bootstrap-actions/blob/master/spark/examples/spark-submit-via-step.md
Comments
Post a Comment