DynamoDB
Daniel Charvát, Denisa Šrámková, Viliam Juríček, Šimon Berka
Introduction to DynamoDB
● Key-value document database developed by Amazon
○ They discovered that 90 % of their operations query a single table
○ SQL database tables was thus mostly redundant
● Fully managed, multimaster, durable database with in-memory caching
● Partition through consistent hashing to spread data across instance nodes
● Size is defined through read and write capacity units
○ Allowed number of operations per second
○ Generally cheaper with less frequent usage
Companies that are using DynamoDB
Source: https://www.featuredcustomers.com/vendor/amazon-dynamodb/customers
Ranking
Source: https://db-engines.com/en/ranking/
Features
Strengths
● Seamless scalability through
automatic instance expansion
● Data backed up to Amazon S3
● Ease of integration with other
AWS services
Drawbacks
● No ACID transactions
○ Although eventual consistency
is almost guaranteed
● Not suitable for large binary
objects.
● Cross-region replicability not
available
Setting up DynamoDB
There are several options:
● Local
○ Windows, Linux, Mac OS (downloadable version)
○ Apache Maven (POM file)
○ Docker
● Web service
1. Sign up to AWS
2. Get an AWS access key
3. Configure credentials
Accessing DynamoDB
Again there are several options:
● AWS Management Console
● AWS Command Line Interface
● DynamoDB API - supports Java, JavaScript, .NET, Node.js, PHP, Python
(AWS SDK called Boto 3), Ruby, C++, Go, Android and iOS
Core components
● a table is a collection of items
● each item is a collection of attributes
● primary keys are used to uniquely identify each item
in a table
● secondary indexes provide more querying flexibility
Other than the primary key, the ‘People’ table is
schemaless, which means that neither the attributes nor
their data types need to be defined beforehand.
Source: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.CoreComponents.html
Core components
The primary key for table ‘Music’ consists of two
attributes (Artist and SongTitle).
Each item in the table must have these two attributes.
The combination of Artist and SongTitle distinguishes
each item in the table from all of the others.
Source: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.CoreComponents.html
Primary key
● must be specified when creating a table
● DynamoDB support two kinds of primary keys:
○ partition key: composed of one attribute (PersonID)
○ partition key and sort key: composed of two attributes (Artist,
SongTitle)
● each primary key attribute must be a scalar (of a data type: string,
number, or binary)
Source: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.CoreComponents.html
Secondary indexes
● we can query the data in the table using an alternate key (in addition to
queries against the primary key)
● two kinds of indexes:
○ global: both partition and sort key can be different from those on the
table
○ local: same partition key as the table, but a different sort key
● default indexes limit per table: 20 global, 5 local
Source: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.CoreComponents.html
Secondary indexes
We can query data items by Artist
(partition key) or by Artist and
SongTitle (partition key and sort key).
If we also wanted to query the data by
Genre and AlbumTitle:
1. Create an index on Genre and
AlbumTitle
2. Query the index
Source: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.CoreComponents.html
RDBMS vs. DynamoDB
Source: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/SQLtoNoSQL.html
Characteristics Relational Database Management
System (RDBMS)
Amazon DynamoDB
Optimal Workloads Ad hoc queries; data warehousing; OLAP Web-scale applications
Data Model Requires a well-defined schema (data is
normalized into tables, rows, and columns).
Schemaless - Can manage structured or
semistructured data.
Performance optimized for storage optimized for compute
Scaling Scale up through faster hardware
tables can be span across multiple hosts in
a distributed system (upper limits on
scalability).
Designed to scale out using distributed clusters of
hardware (No upper limit).
Creating a table - schema example
{
"TableName": "TestCertificates",
"KeySchema": [
{ "AttributeName": "cert_info_link", "KeyType": "HASH" },
{ "AttributeName": "cert_authority", "KeyType": "RANGE" }
],
"GlobalSecondaryIndexes": [
{ "IndexName": "log_type_index",
"KeySchema": [
{
"AttributeName": "log_type",
"KeyType" : "HASH"
},
{
"AttributeName": "cert_authority",
"KeyType" : "RANGE"
}
],
cert_info_link not_valid_before not_valid_afte
r
cert_common_name cert_authority log_type
http://ct.google... 1573257600 1581119999 thesmartlocal0... cPanel, Inc. X509LogEntry
Creating a table - schema example
"Projection": {
"ProjectionType": "ALL"
},
"ProvisionedThroughput" : {
"ReadCapacityUnits": 5,
"WriteCapacityUnits": 5
}
}
],
"AttributeDefinitions": [
{ "AttributeName": "cert_info_link", "AttributeType": "S" },
{ "AttributeName": "cert_authority", "AttributeType": "S" },
{ "AttributeName": "log_type", "AttributeType": "S" }
],
"ProvisionedThroughput": {
"ReadCapacityUnits": 5,
"WriteCapacityUnits": 5
}
}
LIVE DEMO