CODEWITHSUNDEEP
databasesymfonycaching
Antoine Nedelec
Antoine Nedelec

Reputation: 665

Symfony 4 - Cache vs DB solution

I am currently using the cache for my current project, but i'm not sure if it is the right thing to do.

I need to retrieve a lot of data from a web api (nodes that can be picture, node, folder, gallery.... Those nodes will change very often, so I need fast access (loading up to 300-400 element at once). Currently I store them in cache (key as md5 of node_id, so easy to retrieve, and update).

It is working great so far, but if I clear the cache it takes up to 1 minute to create all the cache again.

Should I use a database to store those nodes ? Will it be quicker / slower / same ?

Upvotes: 0

Views: 170

Answers (1)

dbrumann
dbrumann

Reputation: 17166

Your question is very broad and thus hard to answer. Saving 300-400 elements under a cache key sounds problematic to me. You can run into problems where serializing when storing in the cache and deserializing when retrieving the data will cause problems for you. Whenever your cache service is down your app will be practically unusable.

If you already run into problems when clearing/updating the cache you might want to look for an alternative. This might be a database or elasticsearch, advanced cache features like tagged caching could help with preventing you from having to clear the whole cache when part of the information updates. You might also want to use something like the chain provider to store things in multiple caches to prevent the aforementioned problem of an unreachable cache "breaking" your app. You could also look into a pattern that is common with CQRS called a read model.

There are a lot of variables that come into play. If you want to know which one will yield the best results, i.e. which one is quicker, you should do frequent performance tests with realistic data using Symfony's debug toolbar & profilers or a 3rd party service like blackfire.io or tideways. You might also want to do capacity test with a tool like JMeter to ensure those results still hold true, when there are multiple simultaneous users.

Upvotes: 1

Related Questions