AWS Databases and Cache Overview

RDS Solution Architecture

• RDS uses an Elastic Load Balancer for distribution
• Behind the balancer are EC2 instances which can include an Auto Scaling Group (ASG)
• The instances interact with a Read/Write layer
• The Read/Write layer interacts with an Amazon RDS SQL (relational) Database

Amazon Aurora

• Aurora is a proprietary, not open-source, technology from AWS
• It supports PostgreSQL and MySQL
• It's optimized for the AWS cloud, achieving 5x performance increases compared to MySQL on RDS, and 3x over Postgres on RDS.
• Aurora storage automatically extends in 10GB increments up to 64TB.
• Aurora costs more than RDS(approximately 20%) but offers better efficiency.
• It is not included in the free tier.

RDS Deployments: Read Replicas, Multi-AZ

• Read Replicas:
  • Scale the read workload for a database.
  • Up to 15 read replicas are possible
  • Data is written only to the main database
• Multi-AZ:
  • Provides high availability through failover in case of Availability Zone (AZ) outage.
  • Data is written to, and only read from/written to, the primary database.
  • Only one other AZ is allowed as a failover.

RDS Deployments: Multi-Region

• Multi-Region deployments utilize Read Replicas for disaster recovery and improved local performance for global reads.
• Replication is possible across different AWS regions.

Amazon ElastiCache Overview

• ElastiCache provides managed Redis or Memcached in-memory databases for high performance and low latency.
• It reduces the read load on the primary database.
• AWS handles all OS maintenance and other management, such as backups, configuration and failure recovery.

ElastiCache Solution Architecture - Cache

• EC2 instances, possibly in an ASG, interact with ElastiCache.
• ElastiCache is used as a cache between the Elastic Load Balancer and the Amazon RDS SQL relational database.
• Requests are directed first to ElastiCache for a fast read/write operation.
• Slow read/write requests from ElastiCache are then directed to Amazon RDS.

DynamoDB

• A fully managed, high availability database with replication across 3 Availability Zones.
• It's a NoSQL database, not relational.
• It scales to massive workloads through a distributed serverless architecture.
• Delivers millions of requests per second, supporting data volumes up to hundreds of terabytes.
• Its single-digit millisecond latency offers extremely low retrieval latency.
• Integrated with AWS Identity and Access Management (IAM) for secure authorization.
• It’s cost-effective and auto-scalable.

DynamoDB - Type of Data

• DynamoDB is a key-value database.
• Data is organized using primary and partition keys.
• Data is structured using schemas, and attributes must be defined.

DynamoDB Accelerator - DAX

• A fully managed, in-memory cache designed for DynamoDB.
• It improves performance significantly, reducing latency from single digits to microseconds.
• Provides scalability and high availability.
• Designed for use only with DynamoDB, unlike ElastiCache.

Redshift Overview

• Redshift is an analytic database service based on PostgreSQL, though not for OLTP.
• It's used for online analytic processing (OLAP).
• Data loading happens periodically (hourly rather than every second).
• Superior performance compared to other data warehouses, supporting petabytes of data.
• It uses columnar storage for efficiency.
• It features massively parallel query execution (MPP) for high availability.
• Its pay-per-use model is based on provisioned instances.
• Supports standard SQL queries and integrates with business intelligence (BI) tools like QuickSight and Tableau.

Amazon EMR

• EMR stands for Elastic MapReduce.
• Creating Hadoop clusters for analyzing large datasets.
• It supports various tools and technologies.
• It handles the provisioning and setup of clusters, and offers auto-scaling and Spot Instances.
• Used for big data processing, machine learning, and web indexing.

Athena Overview

• A fully serverless database, with SQL capabilities, used for querying data in Amazon S3.
• Querying is pay-per-query.
• Results are stored in S3 and secured through AWS Identity and Access Management (IAM).
• Used for ad-hoc SQL queries and serverless queries on S3, including for log analytics.

Amazon QuickSight

• Amazon QuickSight is a serverless machine learning-powered business intelligence service.
• It generates interactive dashboards and visualizations.
• It's fast, automatically scalable, and priced per session.
• It provides diverse use cases, including business analysis, visualizations, ad-hoc analysis and extracting business insights using data.
• It integrates with various AWS services including RDS, Aurora, Athena, Redshift, and S3.

DocumentDB

• Aurora is an implementation of Postgres/MySQL…
• DocumentDB is a NoSQL database similar to MongoDB (stores, queries and indexes JSON data).
• It's fully managed, highly available via replication across 3 Availability Zones.
• Auto-scaling to handle millions of requests per second.
• Uses 10GB increments for storage extending up to 64 TB.

Amazon Neptune

• Amazon Neptune is a fully managed graph database.
• Suitable for datasets like social networks.
• Highly available in 3 Availability Zones (with 15 read replicas).
• Processes billions of relationships with millisecond latency.
• Suitable for knowledge graphs, fraud detection, recommendations, and more.

Amazon QLDB

• QLDB ("Quantum Ledger Database") is a ledger, essentially a log, for financial transactions.
• A completely managed, serverless, and highly available service that provides ledger storage and change history recording that is cryptographically verifiable.
• Immutable journal ensures that data cannot be corrupted, removed or modified, and offers better data integrity.
• Replicates across 3 zones, suitable for financial applications and systems requiring data integrity.
• Uses an SQL-based query language.

Amazon Managed Blockchain

• Blockchain technology lets many parties conduct transactions without a central authority.
• Amazon Managed Blockchain is a service managing this infrastructure.
• Facilitates creation of private and public blockchain networks using well-known blockchain frameworks (like Hyperledger Fabric and Ethereum).

AWS Glue

• A managed extract, transform, and load (ETL) service.
• Prepares and transforms data for analytics.
• It's serverless.
• Features a Data Catalog cataloging datasets.
• Can be used with services like Athena, Redshift, and EMR.

DMS - Database Migration Service

• Quickly and securely migrates databases to AWS.
• Maintains Availability of the source database during the migration.
• Supports homogeneous and heterogeneous migrations (e.g., Oracle to Oracle, SQL Server to Aurora).

Databases & Analytics Summary in AWS

• Summary details for different types of databases & analytics options available (e.g., relational, in-memory, key-value, data warehousing)

Other Compute Section

• [No specific details]

What is Docker?

• Docker is a software platform for deploying applications.
• Applications are packaged into containers, ensuring consistency across different systems.
• The application runs equally regardless of the underlying OS. Offers no compatibility issues.
• Docker is simpler to implement and maintain than traditional virtual machines, and it supports numerous technologies and languages.
• Scalable; containers can be easily adjusted up or down.

Docker on an OS

• Docker containers run on servers.

Where Docker Images are Stored

• Docker images are stored in repositories.
• Public repositories (like Docker Hub) provide standard image access.
• AWS ECR (Elastic Container Registry) stands as private repository option.

Docker versus Virtual Machines

• Docker is similar to virtualization, with containers sharing resources with the host server.

ECS

• ECS (Elastic Container Service) automates and simplifies the process of deploying Docker containers.
• ECS handles the provisioning, operations and management of the EC2 instances required to run the containers.
• Allows easier integration with other AWS services (e.g., Application Load Balancer).

Fargate

• Fargate is a serverless compute option for running containers within AWS.
• Fargate eliminates the need to handle the underlying EC2 infrastructure for running Docker containers.
• The AWS infrastructure manages the underlying server resources which eliminates the burden of operations and scaling of the virtual server, making it suitable for quick setup.

ECR

• ECR (Elastic Container Registry) is an AWS service that acts as a private repository for Docker images.
• ECR manages the storage and retrieval of your Docker images, facilitating use of Docker containers in other AWS services.

This quiz covers the primary functions and features of AWS databases and caching solutions, including ElastiCache, DynamoDB, and Amazon RDS. Test your understanding of how these services differ, their performance enhancements, and unique characteristics. Perfect for anyone looking to deepen their knowledge of AWS architecture.