diff options
author | Peter Parente <pparent@us.ibm.com> | 2015-04-09 06:37:20 -0400 |
---|---|---|
committer | Sean Owen <sowen@cloudera.com> | 2015-04-09 06:37:20 -0400 |
commit | b9c51c04932efeeda790752276078314db440634 (patch) | |
tree | 715a45df8756fdd26f068a3154ae9d8750bcbd92 /docs/cluster-overview.md | |
parent | 2fe0a1aaeebbf7f60bd4130847d738c29f1e3d53 (diff) | |
download | spark-b9c51c04932efeeda790752276078314db440634.tar.gz spark-b9c51c04932efeeda790752276078314db440634.tar.bz2 spark-b9c51c04932efeeda790752276078314db440634.zip |
[SPARK-6343] Doc driver-worker network reqs
Attempt at making the driver-worker networking requirement more explicit and up-front in the documentation (see https://issues.apache.org/jira/browse/SPARK-6343).
Update cluster overview diagram to show connections from workers to driver. Add a bullet below about how driver listens / accepts connections from workers.
Author: Peter Parente <pparent@us.ibm.com>
Closes #5382 from parente/SPARK-6343 and squashes the following commits:
0b2fb9d [Peter Parente] [SPARK-6343] Doc driver-worker network reqs
Diffstat (limited to 'docs/cluster-overview.md')
-rw-r--r-- | docs/cluster-overview.md | 6 |
1 files changed, 5 insertions, 1 deletions
diff --git a/docs/cluster-overview.md b/docs/cluster-overview.md index 6a75d5c457..7079de546e 100644 --- a/docs/cluster-overview.md +++ b/docs/cluster-overview.md @@ -33,7 +33,11 @@ There are several useful things to note about this architecture: 2. Spark is agnostic to the underlying cluster manager. As long as it can acquire executor processes, and these communicate with each other, it is relatively easy to run it even on a cluster manager that also supports other applications (e.g. Mesos/YARN). -3. Because the driver schedules tasks on the cluster, it should be run close to the worker +3. The driver program must listen for and accept incoming connections from its executors throughout + its lifetime (e.g., see [spark.driver.port and spark.fileserver.port in the network config + section](configuration.html#networking)). As such, the driver program must be network + addressable from the worker nodes. +4. Because the driver schedules tasks on the cluster, it should be run close to the worker nodes, preferably on the same local area network. If you'd like to send requests to the cluster remotely, it's better to open an RPC to the driver and have it submit operations from nearby than to run a driver far away from the worker nodes. |