Github user maropu commented on the pull request:
https://github.com/apache/spark/pull/10346#issuecomment-165349453
I did quick benchmarks for large broadcasts;
- aws m4.x4large x 4, 4 works in a host
- elapsed time:
-- w/opt.: 6.887943434s, w/o opt.: 11.738593435s
- received network bytes in a host:
-- w/opt.: 14782.93kB/s, w/o opt.: 35394.53kB/s
```
val bv = sc.broadcast((0 until 50000000).toArray)
val parts = sc.parallelize((1 to 16), 16).cache
def timer[R](block: => R): R = {
val t0 = System.nanoTime()
val result = block
val t1 = System.nanoTime()
println("Elapsed time: " + ((t1 - t0 + 0.0) / 1000000000.0)+ "s")
result
}
timer {
parts.map(_ => bv.value.size).foreach(_ => {})
}
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
