Caching Architecture

Embedded Cache

• most common

• Copy of data (from db, expensive io call, or expensive calculation) put in memory

• Common with cpu and l4 cache

• Within applications, this is just a map/dictionary

Client-Service Cache

• cache is placed outside of application and is placed between app and the other system

• Proxy

• Having multiple apps or replicas that talk to the cache instead

Distributed Cache

• Instead of having one cache, will have multiple caches, to increase avialability of cache, or locate cache closer to consumer

• low latency high volume

Cloud Cache or CSP Managed Cache

• distributed cache that is managed by cloud provider

• Dont need to manage it

• improve scalability

Reverse-Proxy Cache

• responses cached

• GET requests

• cache stored inside the proxy

• avoids bombarding servers

Side-car Cache

• The cache is attached to container running in the pod

• cache per replica

• lifecycle tied to the pod

• low latency

• app and data isolation

Reverse-Proxy Side-car Cache