DevOpsBootcampUPES

DynamoDB

Create a table

1. Open the DynamoDB console at https://console.aws.amazon.com/dynamodb/
2. In the navigation pane on the left side of the console, choose Dashboard.
3. On the right side of the console, choose Create Table.
4. Enter the table details as follows:
   1. For the table name, enter a unique table name.
   2. For the partition key, enter Artist.
   3. Enter SongTitle as the sort key.
   4. Choose Customize settings.
   5. On Read/write capacity settings choose Provisioned mode with autoscale capacity with a minimum capacity of 1 and maximum of 10.
5. Choose Create to create the table.

Write and read data

1. On DynamoDB web console page, choose PartiQL editor on the left side menu.
2. The following example creates several new items in the <table-name> table. The PartiQL is a SQL-compatible query language for DynamoDB.

INSERT INTO "<table-name>" VALUE {'Artist':'No One You Know','SongTitle':'Call Me Today', 'AlbumTitle':'Somewhat Famous', 'Awards':'1'}

INSERT INTO "<table-name>" VALUE {'Artist':'No One You Know','SongTitle':'Howdy', 'AlbumTitle':'Somewhat Famous', 'Awards':'2'}

INSERT INTO "<table-name>" VALUE {'Artist':'Acme Band','SongTitle':'Happy Day', 'AlbumTitle':'Songs About Life', 'Awards':'10'}
                            
INSERT INTO "<table-name>" VALUE {'Artist':'Acme Band','SongTitle':'PartiQL Rocks', 'AlbumTitle':'Another Album Title', 'Awards':'8'}

Query the data by

SELECT * FROM "<table-name>" WHERE Artist='Acme Band' AND SongTitle='Happy Day'

Spot check

From your local machine, write some more items using the aws dynamodb put-item command. Use the aws dynamodb get-item to get some item.

Solution

Write

aws dynamodb put-item --table-name Music --item '{"Artist": {"S": "No One You Know"}, "SongTitle": {"S": "Call Me Today"}, "AlbumTitle": {"S": "Somewhat Famous"}, "Awards": {"N": "1"}}'

Get

aws dynamodb get-item --table-name Music --key '{ "Artist": {"S": "Acme Band"}, "SongTitle": {"S": "Happy Day"}}'

Create and query a global secondary index

1. In the navigation pane on the left side of the console, choose Tables.
2. Choose your table from the table list.
3. Choose the Indexes tab for your table.
4. Choose Create index.
5. For the Partition key, enter AlbumTitle.
6. For Index name, enter AlbumTitle-index.

7. Leave the other settings on their default values and choose Create index.

8. You query the global secondary index through PartiQL by using the Select statement and providing the index name

   SELECT * FROM "<table-name>"."AlbumTitle-index" WHERE AlbumTitle='Somewhat Famous'
   

Self-check questions

Enter the interactive self-check page

Exercises

:pencil2: Database migration

Database migration is the process of transferring data and schema from one database to another, typically when transitioning to a new system or version.

Zero downtime database migration refers to the seamless migration of a database without any interruption or impact on the availability of the system. It involves careful planning, replication, and synchronization techniques to ensure continuous service while transitioning to the new database environment.

1. Start mongo docker container.
2. Start the script under db_migration/init.py. The script should write new data to your running mongo every second. Keep this script running while you work on your next step.
3. Based in the script in db_migration/init.py, create another script called mirror_db.py such that every data that is written to mongo will be written to dynamoDB table (create this table with dataId as a partition key).
4. (Optional) Use try...except to catch and handle the case where the write operation for either mongo or dynamo is failed.
5. Stop the running script from step 2, keep the timestamp of the last document that was written to mongo.
6. Start the mirror_db.py script. From now on, every piece of data is being written to both mongo and dynamo. You only need to create to the data written before you started the mirror script.
7. Write a script that copies all the documents with timestamp smaller than or equal to the timestamp you kept from step 5. This script should be run only once.

After a while… stop the mirror_db.py script and check that the data stored in both DBs is identical (you may write some python script for that).

:pencil2: Point-in-time recovery for DynamoDB

Restore your table to how it was looking like a few minutes ago.

https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/PointInTimeRecovery.Tutorial.html#restoretabletopointintime_console

This site is open source. Improve this page.