Exercise 3 – Partitions

In this exercise, you will:

• Experiment with partitions

Steps

NOTE: Be sure Apache Cassandra™ is running before doing these exercises. You can check by running dsetool on the command line:

/home/ubuntu/node/bin/dsetool status

1) Start the CQL command shell at the command line:

/home/ubuntu/node/resources/cassandra/bin/cqlsh

2) Switch to the killrvideo keyspace via the USE command:

USE killrvideo;

3) Execute the following command to view the metadata for the videos table you created earlier.

DESCRIBE TABLE videos;

• What is the partition key?
• How many partitions are in this table?

4) Execute the following query to view the partitioner token value for each video id.

SELECT token(video_id), video_idFROM videos;

5) Exit cqlsh and use the following command to inspect the file named /home/ubuntu/labwork/data-files/videos-by-tag.csv:

cat /home/ubuntu/labwork/data-files/videos-by-tag.csv

6) Restart cqlsh and switch to the killrvideo keyspace.

7) Your mission, should you choose to accept it, is to write a CREATE TABLE statement that will store this data partitioned by tags. With this given data set, there should be two partitions, one for each tag. Call your table videos_by_tag.

8) Execute the following COPY command to import the videos-by-tag.csv data.
COPY videos_by_tag(tag, video_id, added_date, title) FROM '/home/ubuntu/labwork/data-files/videos-by-tag.csv' WITH HEADER = TRUE;

9) Verify CQL imported your data correctly by writing a SELECT * command.
SELECT * FROM videos_by_tag;

• Note that if the table only contains 2 records, the primary key may be wrong -containing only the tag field -be sure to include the video_id as a cluster column.

10) Write a SELECT statement to retrieve all rows tagged with cassandra.

11) Now, find all videos tagged with datastax (similar to the previous query).

12) Finally, write a query to retrieve the video having a title of Cassandra Intro.