Install Spark on the cluster: Difference between revisions
From info319
| Line 53: | Line 53: | ||
mkdir -p ~/volume/python | mkdir -p ~/volume/python | ||
cd ~/volume/python | cd ~/volume/python | ||
sudo apt install python3-pip python3-dev python3-venv | sudo apt install python3-pip python3-dev python3-venv | ||
python3 -m venv venv | python3 -m venv venv | ||
. venv/bin/activate | . venv/bin/activate | ||
python3 -m pip install --upgrade pip | python3 -m pip install --upgrade pip | ||
pip install findspark | pip install tweepy # kafka-python afinn | ||
# pip install findspark | |||
# because SPARK_HOME is set, findspark is not needed | # because SPARK_HOME is set, findspark is not needed | ||
sudo apt install emacs | |||
# edit exercise1.py, first to read form big file, then from hdfs | |||
--> | --> | ||
'''Task:''' Run ''exercise1.py'' from Exercise 1, for example with | '''Task:''' Run ''exercise1.py'' from Exercise 1, for example with ''spark-submit exercise1.py''. A few tips: | ||
* Use the large dataset in ''tweets_id_text_100000.jl'' (i.e., '''small_dataset = False'''). | * Use the large dataset in ''tweets_id_text_100000.jl'' (i.e., '''small_dataset = False'''). | ||
* Because SPARK_HOME is set, you do not need ''findspark''. | * Because SPARK_HOME is set, you do not need ''findspark''. | ||
* | * When you run on top of YARN, Spark expects to input files from HDFS, not from the regular file system. | ||
* Use the HDFS and YARN web UIs to | * Use the HDFS and YARN web UIs to check what is going on. | ||
* The default replication factors and other settings are not clever. Don't worry about that now: they are enough to get you started. | * The default replication factors and other settings are not clever. Don't worry about that now: they are enough to get you started. | ||
'''Task:''' Run the full Twitter pipeline from Exercise 3. | '''Task:''' Run the full Twitter pipeline from Exercise 3. Tips: | ||
* You need to create virtual environment to download packages, specifically ''tweepy'', ''kafka-python'', and ''afinn''. | |||
* You need to create ''keys/bearer_token'' with restricted access. | |||
* With the default settings, it can take time before the final join is written to console. Be patient... | |||
* Start-up sequence from ''spark-driver'': | |||
# assuming zookeeper and kafka run already | |||
# create kafka topics | |||
kafka-topics.sh --create --topic tweets --bootstrap-server localhost:9092 | |||
kafka-topics.sh --create --topic media --bootstrap-server localhost:9092 | |||
kafka-topics.sh --create --topic photos --bootstrap-server localhost:9092 | |||
# optional monitoring | |||
kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic tweets --from-beginning & | |||
kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic media --from-beginning & | |||
kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic photos --from-beginning & | |||
# main programs | |||
python twitterpipe_tweet_harvester.py # run this repeatedly or modify to reconnect after sleeping | |||
python twitterpipe_media_harvester.py & # this one can run in background | |||
spark-submit --packages org.apache.spark:spark-sql-kafka-0-10_2.12:3.3.0 twitterpipe_tweet_pipeline.py & | |||
Because you no longer use ''findspark'', you need to use the '''--packages''' option with '''spark-submit'''. Make sure you use exactly the same package as you did with ''findspark.add_packages(...)'''. | |||
== Web UIs == | == Web UIs == | ||
While spark is running, you can attempt to access Spark's web UI at http://158.39.201.197:4040 (assuming that 158.39.201.197 is the IPv4 address of ''spark-driver''). But when you run on top of YARN it just attempts to redirect to YARN's web UI at http://158.39.201.197:8088 , which you have accessed already. | While spark is running, you can attempt to access Spark's web UI at http://158.39.201.197:4040 (assuming that 158.39.201.197 is the IPv4 address of ''spark-driver''). But when you run on top of YARN it just attempts to redirect to YARN's web UI at http://158.39.201.197:8088 , which you have accessed already. | ||
Revision as of 17:42, 17 October 2022
Install Spark on the cluster
Install Spark
Go to Apache Spark Downloads. Download and unpack a recent Spark binary. For example, on each instance:
cd ~/volume wget https://dlcdn.apache.org/spark/spark-3.3.0/spark-3.3.0-bin-hadoop3.tgz tar xzvf spark-3.3.0-bin-hadoop3.tgz rm spark-3.3.0-bin-hadoop3.tgz ln -fns spark-3.3.0-bin-hadoop3 spark
Set $SPARK_HOME and add Spark binaries to path:
export SPARK_HOME=/home/ubuntu/volume/spark
export PATH=${PATH}:${SPARK_HOME}/bin:${SPARK_HOME}/sbin
cp ~/.bashrc ~/.bashrc-bkp-$(date --iso-8601=minutes --utc)
echo "export SPARK_HOME=/home/ubuntu/volume/spark" >> ~/.bashrc
echo "export PATH=\${PATH}:\${SPARK_HOME}/bin:\${SPARK_HOME}/sbin" >> ~/.bashrc
On your local machine, create the file spark-defaults.conf:
spark.master yarn spark.driver.memory 512m spark.yarn.am.memory 512m spark.executor.memory 512m
From the local machine, upload spark-defaults.conf to each instance:
scp spark-defaults.conf spark-driver:volume/spark/conf scp spark-defaults.conf spark-worker-1:volume/spark/conf ...
Test run Spark
On spark-driver, start Spark with:
SPARK_PUBLIC_DNS=$MASTER_NODE pyspark
SPARK_PUBLIC_DNS is the public IP address that Spark's web UI listens to at port 4040. To set it permanently:
export SPARK_PUBLIC_DNS=$MASTER_NODE echo "export SPARK_PUBLIC_DNS=$MASTER_NODE" >> ~/.bashrc
Create a test program, e.g., spark-test.py:
from pyspark.sql import SparkSession
spark = SparkSession.builder \
.master("local") \
.appName("Word Count") \
.config("spark.some.config.option", "some-value") \
.getOrCreate()
print('*** The result row is',
spark.range(5000).where("id > 500").selectExpr("sum(id)").collect(),
'***')
Test run with:
spark-submit spark-test.py
Task: Run exercise1.py from Exercise 1, for example with spark-submit exercise1.py. A few tips:
- Use the large dataset in tweets_id_text_100000.jl (i.e., small_dataset = False).
- Because SPARK_HOME is set, you do not need findspark.
- When you run on top of YARN, Spark expects to input files from HDFS, not from the regular file system.
- Use the HDFS and YARN web UIs to check what is going on.
- The default replication factors and other settings are not clever. Don't worry about that now: they are enough to get you started.
Task: Run the full Twitter pipeline from Exercise 3. Tips:
- You need to create virtual environment to download packages, specifically tweepy, kafka-python, and afinn.
- You need to create keys/bearer_token with restricted access.
- With the default settings, it can take time before the final join is written to console. Be patient...
- Start-up sequence from spark-driver:
# assuming zookeeper and kafka run already # create kafka topics kafka-topics.sh --create --topic tweets --bootstrap-server localhost:9092 kafka-topics.sh --create --topic media --bootstrap-server localhost:9092 kafka-topics.sh --create --topic photos --bootstrap-server localhost:9092 # optional monitoring kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic tweets --from-beginning & kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic media --from-beginning & kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic photos --from-beginning & # main programs python twitterpipe_tweet_harvester.py # run this repeatedly or modify to reconnect after sleeping python twitterpipe_media_harvester.py & # this one can run in background spark-submit --packages org.apache.spark:spark-sql-kafka-0-10_2.12:3.3.0 twitterpipe_tweet_pipeline.py &
Because you no longer use findspark, you need to use the --packages' option with spark-submit. Make sure you use exactly the same package as you did with findspark.add_packages(...).
Web UIs
While spark is running, you can attempt to access Spark's web UI at http://158.39.201.197:4040 (assuming that 158.39.201.197 is the IPv4 address of spark-driver). But when you run on top of YARN it just attempts to redirect to YARN's web UI at http://158.39.201.197:8088 , which you have accessed already.
