Caching

• Performance of the application can be improved.
• Scalability of the system by reducing the load on key components of the application
• Move frequently accessed data closer to the application
• Faster response time
• Resilience by having a read-only copy of data make it more stable and responsive

Caching should be considered for

• Repeatedly accessed data, especially if that data remains unchanged
• Data source performance
• Data contention, like when multiple processes competing for the same data
• Physical location, to reduce network latency

Caching Patterns

• Cache-aside Pattern stores the most frequently accessed data in the cache, so it's quickly available without first having to query the database.
• Content Cache Pattern will store static content like images, stylesheets, etc., and will reduce the load on the server.
• User Session Caching Pattern is used to maintain the application state between sessions.

Advanced Caching Patterns

• Job and Message Queuing to handle long-running tasks or operations that can negatively impact the application's performance. So offload them to queues where they are run in sequence.
• Distributed Transactions for a group of commands that must be run against a back-end datastore that acts as a single atomic operation.

More details about Caching Patterns - snippset.

Things to consider to estimate the size of the cache:

• Number of concurrent cached objects
• Size of cache objects
• Number of cache requests
• Cache expiration policy

Caching is a mechanism to store frequently accessed data in a data store temporarily and retrieve this data for subsequent requests instead of extracting it from the original data source. This process improves the performance and availability of an application. Reading data from the database may be slower if it needs to execute complex queries.