DEVOPS
BOOTCAMP
Databases
Databases are used to persist data
create
read
update
delete
Users can create, read, update or delete entries in a database
Also called CRUD operations
Database in software development process - 1
Developers need database for local development
Connect to a development database to develop the new features
Connect to a test database to test the new feature with realistic data
...
Database in software development process - 2
2 ways for developers to work with DB
Each developer installs DB locally
can't mess up someone else's test DB data
Option 1 Each developer has own DB with own
test data start from empty DB | need realistic test data
install and setup DB for EVERY developer
Ideal solution: have both!
Option 2 Shared DB hosted remotely no need for local installation | start coding right away
test data available right from the start
can't play around without affecting others!
How does the application
Configure Database Connection - 1 talk to the database?
The DB connection is configured in the application's code
Each programming language has a library/module for DB connection
WHICH database to talk to HOW to AUTHENTICATE with that DB
You have to tell the library:
= =
Database Endpoint Credentials
Example: Java Code connecting to a Mysql database
Configure Database Connection - 2
Better: Do not hardcode database Best Practice: DON'T check in
connection data in source code credentials in code
Define in 1 place as environment variables
Configure DB connection from OUTSIDE
1. Define only variables in code
2. Set endpoint and credentials from outside for each environment
(for example dev, test, prod) DEV TEST PROD
3. Depending on environment (dev, test, prod) connects to different
DB!
Configure Database Connection - 3
The way to do this, is to pass environment variables on application start-up
1) From Command Line
How to pass envrionment variables:
2) Configure in code editor
3) Use properties/configuration files
The best option is to pass them via a properties or configuration file
Configure Database Connection - 4.1
In a Java/Spring application you can define the values in a
Spring properties file:
3 property files for each environment
Values defined in a properties file
Usage in Java code
Configure Database Connection - 4.2
Nodejs application with configuration file:
Using a configuration file is:
cleaner
more transparent
Databases in Production
Application connects to database
This means database needs to be available , when
application starts
So before application is deployed, we need to install
and configure database PROD PROD
Data is important so we need to secure them: This may be done by:
1. Replicate database
2. Do regular backups
3. Make sure it performs under high load System Database DevOps
Administrator Engineer Engineer
Database Types - 1
Key Value Databases Relational Databases Graph Databases
Wide Column Databases Document Databases Search Databases
Many different database types available
They have different characteristics for different use cases
Each of them has benefits and disadvantages
Database Types - 2 Key Value Databases
Examples:
Redis Memcached etcd from Kubernetes
Characteristics Best for:
very fast
Data is saved in key-value pairs Caching
unique key Message Queue limited storage
no joins no primary database
in-memory
Database Types - 3 Wide Column Databases
Examples:
Characteristics Best for:
scalable
A two-dimensional key-value store Time-Series
no joins
Data is saved in tables, rows and IoT-Records
columns limited compared to relational DB
Names and format can vary from
no primary DB
row to row (schema-less)
Queries similar to SQL
Database Types - 4 Document Databases
Examples:
Collection:
MongoDB DynamoDB CouchDB
Characteristics Best for:
faster to read
Data is stored as JSON-like Mobile Apps
easy to get started
documents Game Apps
Schema-less CMS primary database
No joins Most Apps
slower writes
Denormalized
graphs
Database Types - 5.1 Relational Databases
Examples:
Characteristics Best for: ACID
Data is stored in Tables, Rows and structured data easy to get started
Columns
primary database
Schema (schema and data types need
to be created first) difficult to scale
Normalized to avoid duplicated data unstructured data
Most used in industry
Database Types - 5.2 Relational Databases
ACID = Atomicity, Consistency, Isolation, Durability
No half-changes are updated in database
Either ALL changes get applied or NON
SQL vs No-SQL
Because relational or SQL databases are so popular,
often we talk about SQL vs No-SQL, meaning all the other
database types, which are schema-less
Database Types - 6 Graph Databases
Examples:
Neo4j Dgraph
Characteristics Best for:
easier to query
Data is stored as Nodes and Relationships Graphs
instead of tables or documents Patterns
Directly connect Entities
Edges are the Relationships
Database Types - 7 Search-Engine Databases
Characteristics
Dedicated to the search of data
Examples:
Optimized for dealing with data
ElasticSearch Solr
Full-text search in efficient and fast way
Similar to document-orientated database
Usage of indexes
Wrap Up
Choose database type based on your application needs
Or Combinations
For most of data Relational
For example: To handle fast search Search Engine
Cache Key Value
