Spark: what's the advantages of having multipl

Question

I am running my job on AWS-EMR cluster. It is a 40 nodes cluster using cr1.8xlarge instances. Each cr1.8xlarge has 240G memory and 32 cores. I can run with the following config:

--driver-memory 180g --driver-cores 26 --executor-memory 180g --executor-cores 26 --num-executors 40 --conf spark.default.parallelism=4000

or

--driver-memory 180g --driver-cores 26 --executor-memory 90g --executor-cores 13 --num-executors 80 --conf spark.default.parallelism=4000

Since from the job-tracker website, the number of tasks running simultaneously is mainly just the number of cores (cpu) available. So I am wondering is there any advantages or specific scenarios that we want to have more than one executor per node?

Thanks!

Answer 1

Yes, there are advantages of running multiple executors per node - especially on large instances like yours. I recommend that you read this blog post from Cloudera.

A snippet of the post that would be of particular interest to you:

To hopefully make all of this a little more concrete, here’s a worked example of configuring a Spark app to use as much of the cluster as possible: Imagine a cluster with six nodes running NodeManagers, each equipped with 16 cores and 64GB of memory. The NodeManager capacities, yarn.nodemanager.resource.memory-mb and yarn.nodemanager.resource.cpu-vcores, should probably be set to 63 * 1024 = 64512 (megabytes) and 15 respectively. We avoid allocating 100% of the resources to YARN containers because the node needs some resources to run the OS and Hadoop daemons. In this case, we leave a gigabyte and a core for these system processes. Cloudera Manager helps by accounting for these and configuring these YARN properties automatically.

The likely first impulse would be to use --num-executors 6 --executor-cores 15 --executor-memory 63G. However, this is the wrong approach because:

63GB + the executor memory overhead won’t fit within the 63GB capacity of the NodeManagers. The application master will take up a core on one of the nodes, meaning that there won’t be room for a 15-core executor on that node. 15 cores per executor can lead to bad HDFS I/O throughput. A better option would be to use --num-executors 17 --executor-cores 5 --executor-memory 19G. Why?

This config results in three executors on all nodes except for the one with the AM, which will have two executors. --executor-memory was derived as (63/3 executors per node) = 21. 21 * 0.07 = 1.47. 21 – 1.47 ~ 19.