Understanding Databricks & Apache Spark Performance Tuning: Lesson 01 - Spark Architecture
Ғылым және технология
A popular interview question and a critical topic for all Databricks and Spark developers, how do you tune and optimize Spark queries? This video provides a conceptual understanding of where things can go wrong as a starting point to understanding performance tuning and optimization.
Support me on Patreon
www.patreon.com/bePatron?u=63...
Slides
github.com/bcafferky/shared/b...
Пікірлер: 17
Thanks, very nice video 😉
5:54, better comedian than half the comedians in the world
Waiting for next video
Thank you so much. Excellent video.
@BryanCafferky
4 ай бұрын
Thanks
Thanks Sir ❤
I don't know if it's always true, but I've recently discovered that python can be significantly faster that some spark SQL operations such as joins. I'll check, but do you have a video about monitoring cluster performance? I kind of miss the ganglia ui. Thanks Bryan. As always, you're a great teacher and explainer of things. ❤
@BryanCafferky
2 ай бұрын
This may help learn.microsoft.com/en-us/azure/databricks/compute/cluster-metrics
is it possible to run spark nodes on already concurrent HDFS?
@BryanCafferky
2 ай бұрын
Park can read from HDFS. Is that your question?
@Andy-rw4hn
2 ай бұрын
@@BryanCafferky yes. I was not aware of PXF HDFS connector hdfs:parquet. Thank you.
11:50 I actually thought that the data for the query in the black box does not have to be distributed/indexed by City and the select/group-by can be easily made concurrent by itself
@BryanCafferky
2 ай бұрын
I am oversimplifying but when you request joins or aggregations, you trigger a shuffle which the documentation explains as reordering the data over the cluster nodes to, for example, co-locate data keys from the joined tables. See www.talend.com/resources/intro-apache-spark-partitioning/
@BryanCafferky
2 ай бұрын
i'm trying to find more detailed info on this. Thanks.
@Andy-rw4hn
2 ай бұрын
@@BryanCafferky Thank you
There's also Spark Rapids (GPU)
@BryanCafferky
Ай бұрын
Cool. Did not know that. Here's how to set that up on Databricks. Thanks! docs.nvidia.com/spark-rapids/user-guide/23.12.2/getting-started/databricks.html