click below
click below
Normal Size Small Size show me how
Databases
Databases and Analytics on AWS
| Question | Answer |
|---|---|
| DynamoDB | Fast and flexible Non-relational database service (NoSQL) Any scale; scales with no down time Can handle millions of requests per second Consistent performance In memory cache with 10x performance improvement - DynamoDB Accelerator (DAX) Standard/IA |
| Relational Database Service (RDS) | Managed service - easy to set up, operate and scale a relational db in the cloud SQL capability Suited for online transaction processing (OLTP) |
| Database Migration Service | Migrate a database to AWS but continue using it during the migration - source db remains fully operational during migration to minimise downtime Quick Secure Migrate between homogeneous or heterogeneous database types |
| ElastiCache | Enables deployment and running of Memcached or Redis protocol-compliant server nodes In-memory database Reduce load off databases for read intensive workloads High performance Low latency |
| Glue | Serverless, fully managed ETL service for preparing data so it can be loaded for analytics |
| Managed Blockchain | Fully managed service for creation/maintenance of scalable blockchain networks Open source frameworks: Hyperledger Fabric and Ethereum Decentralised - allows multiple partners to execute transactions without the need of a trusted, centralised authority |
| Redshift | Fully managed, petabyte-scale data warehouse service OLAP optimised (online analytical processing) Can use SQL NOT used for OLTP Highly available due to parallel query execution |
| Aurora | Proprietary to AWS Relational database, using MySQL and PostgreSQL Cloud-optimised High performance and high availability |
| Athena | Interactive query service that makes it easy to analyse data in S3 using standard SQL Always serverless - so no infrastructure to manage Pay only for queries |
| Elastic Map Reduce (EMR) | Web service enabling users to easily/cost-effectively process vast amounts of data Uses Hadoop clusters (Big Data) in EC2 instances working in parallel Includes spot instances and autoscaling |
| Glue Data Catalogue | Central repository to store structural and operational metadata for data assets in AWS Glue Store table definition, physical location, business attributes, usage tracking |
| Quantum Ledger Database (QLDB) | Fully managed Immutable ledger database Transparent, cryptographically verifiable transaction log owned by a central trusted authority Tracks every application data change |
| Neptune | Fully managed graph database service Build and run applications that work with highly connected datasets Useful for knowledge graphs, fraud detection, recommendations engines, social networking |
| QuickSight | Use ML-powered business intelligence to create interactive dashboards |
| DocumentDB | Fully managed Has MongoDB compatibility Fast, scalable, highly available Document database service Good for JSON |
| RDS deployments | Multi-AZ: high availability Read Replicas: scalability Multi-Region: disaster recovery and local performance |
| What is a database? | Storing data in a structured way, with the ability to define indexes to improve query efficiency and define relationships between datasets Can only scale vertically Optimised for a purpose and come with niche features |