Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

How does Spark use Netty?

I know that Storm now runs on Netty for communication betwen nodes?

Does Apache Spark also use Netty? If so, in what way?

like image 887
pphanireddy Avatar asked Dec 08 '22 08:12

pphanireddy


1 Answers

Spark uses Akka Actor for RPC and messaging, which in turn uses Netty.

Also, for moving bulk data, Netty is used.

  • For shuffle data, Netty can be optionally used. By default, NIO is directly used to do transfer shuffle data.
  • For broadcast data (driver-to-all-worker data transfer), Jetty is used by default.
like image 124
Tathagata Das Avatar answered Jan 21 '23 22:01

Tathagata Das