Closed petro-rudenko closed 4 years ago
It is a bit late, but I would ask @tonycurtis to review this.
@shamisp sure, feel free to review! I'll make another PR if needed. Will update readme when main code part would be merged.
Oops, looked at #6 by mistake, deleted that comment here.
.github/workflows/sparkucx-ci.yml
15: the SparkUCX
README.md
17: Put SparkUCX jar file to $SPARK_UCX_HOME
to -> in
@tonycurtis - feel free to review the entire README.md :) thanks !!!
On Oct 9, 2019, at 10:32 AM, Pavel Shamis (Pasha) notifications@github.com wrote:
@tonycurtis https://github.com/tonycurtis - feel free to review the entire README.md :) thanks !!!
SparkUCX is a high performance ShuffleManager plugin for Apache Spark that uses RDMA and other high performance transport, supported by UCX https://github.com/openucx/ucx#supported-transports when performing Shuffle data transfers in Spark jobs.
This open-source project is developed, maintained and supported by UCF consortium http://www.ucfconsortium.org/.
SparkUCX is a high performance ShuffleManager plugin for Apache Spark, that uses RDMA and other high performance transports that are supported by UCX https://github.com/openucx/ucx#supported-transports, to perform Shuffle data transfers in Spark jobs.
This open-source project is developed, maintained and supported by the UCF consortium http://www.ucfconsortium.org/.
To enable the SparkUCX Shuffle Manager plugin, add the following configuration property to spark (e.g. $SPARK_HOME/conf/spark-defaults.conf):
To enable the SparkUCX Shuffle Manager plugin, add the following configuration property to spark (e.g. in $SPARK_HOME/conf/spark-defaults.conf):
Tony
@tonycurtis - much appreciated !
Updated README information: