Rethink Redis Solution

[mike-webster]

Description

I implemented a redis cache in an attempt to:

1. Reduce the number of overall calls made to spotify
2. Increase page load time

Great in theory... but it's not working as planned.

Technical Details

• The provider I'm using is a heroku dyno. I'm using the free one right now, which has a hard limit of 5MB.
  • I figured it would be better to use this to try to get a feel for what I might actually need, but I'm hitting the ceiling here with just me as a user while I'm developing.
• The actual information I'm storing could be optimized
  • I believe I'm storing the entire response body. This was for simplicity, not because I thought it was optimal.
  • I'm not sure the best way to optimize the data. There will obviously be tradeoffs no matter what. It's probably the optimal speed right now; everything I do to parse the information will make the process take longer - so be smart.
• The keys I'm using are absurdly long
  • I believe it's just parsing the urls being requested, including query strings. This is an issue because some of the requests will include a list of ids, which can make the url very long
  • try to find a way to set custom keys

[mike-webster]

This is what I was using to inspect my redis content