-
Useful Links:
-
Exam Tips:
- Supports any RDS DB or even no DB at all is required.
- Be aware of scenarios where it is need:
- Used to improve database performance by caching query results.
- Used where read heavy workloads with low latency requirements.
- Reduces database workloads, meaning savings.
- Application needs to know to use caching. - Application changes.
- Understand the differences between Redis and Memcached:
- Advanced data structures beyond key value stores, you have to use Redis.
- Vertical scaling = Memcached
- High Availability = Redis
- Data Durability:
- Redis data is not persistent by default, but you can enable the Append-Only File (AOF). This allows the Redis node to write all of the commands that modify the cache data to an append-only file.
- In the case a Redis node is rebooted, the AOF allows the Redis node to have a warm cache with the data intact.
- Disabled by default, enabled by assigning a parameter group with
appendonly
set to yes. - You can use
appendfsync
to determine how often the AOF file is written to.
Memcached | Redis |
---|---|
Simple Data Structures | Advanced Structures |
No replication | Multi-AZ |
Multiple nodes (Sharding) | Replication (Scaled Reads) |
No backups | Backup & restore |
Multi-threaded | Transactions |