33 lines
1.3 KiB
Markdown
33 lines
1.3 KiB
Markdown
---
|
|
tags:
|
|
- AWS
|
|
- databases
|
|
---
|
|
|
|
# Database options
|
|
|
|
Amazon offers numerous database options however they are often named differently
|
|
or are Amazon's own implementation of a certain database type:
|
|
|
|
- **Relational databases**
|
|
|
|
- Amazon offers two core RDSs: **Amazon Aurora** and **Amazon RDS**. Both
|
|
allow you to use MySQL and PostgreSQL as the management implementation. The
|
|
differences come down to which management implmentations are supported,
|
|
number of backups and throughput. For instance Aurora has higher
|
|
performance, more logging and fail-safes.
|
|
|
|
- **NoSQL**
|
|
- Document-based databases
|
|
- The obvious candidate for this kind of database is MongoDB. Amazon doesn't
|
|
offer MongoDB it offers its own implementation which is almost identical
|
|
and compatible with Mongo: **Amazon DocumentDB**. (There is some bitching
|
|
about this from Mongo but they are competitors.)
|
|
- Key-value database
|
|
- A key-value database is basically a hash-table or a map
|
|
- The Amazon implementation of this is **DynamoDB**
|
|
- Confusingly, for NoSQL, a DynamoDB instance comprises "tables" which are
|
|
groupings of keys and values. I guess "table" here is meant in terms of
|
|
"hash table"
|
|
- A cool feature is that you can subscribe to a DynamoDB table and run a
|
|
lambda when a value changes.
|