In-memory Data Stores

Redis

Redis is a data store, useful for simplifying a system, increasing responsiveness, and capable of working in addition to other SQL and NoSQL servers.

Redis is a data structures server, using key-value storage format (like hashes) with support for different types of values. Redis can hold more complex data structures than regular key-value stores.

Advantages of Redis

• It doesn't have to replace elements of the current tech stack, it can be added to the current stack.
• Disk persistent memory makes data retrieval much faster than SQL and NoSQL data stores.

source

Memcached

Memory object caching system.

• speeds up dynamic web applications
  • lightening database loads
• in-memory key-value store for small database and api calls, or page rendering
• allows you to make better use of available cache space from multiple web servers
  • done by combining the cache size of multiple servers into one combined cache

What is it made up of

• client software, which is given a list of available memcached servers
• client-based hashing algorithm which chooses a server based on the "key"
• server software, stores values with their keys into a hash table
• LRU (algorithm), which determines when to discard old memory

Half server, half client logic

• clients choose which server to read and write from
• servers manage storage and implement LRU
• servers are unaware of each other

Memcached vs Redis

Both in memory key/val lookup tables, for high speed/availability.

• both cluster, add in memory objects without needing to reference a data store
• comparable access speeds
• past a single instance, they have different deployed and scaled architectures