'Redis invalidation strategies based on inactivity timeouts

I am using ioredis library in nodeJS to write user sessions into cache and retrieve them when the process is restarted.

I am using polling connection between client and server making a ping to server every 2 seconds. In each of the ping request, I am extending the session timeout by 10 more minutes in the memory.

Since I have the sessions stored on Redis as well, I also require this to be up to date on the cache. I am expecting several tens of thousands of people using the service concurrently. At this rate, I am not sure if updating Redis at this rate is okay.

Is there a programming standard or pattern on how people handle this sort of "timeout based cache invalidation"?

node.jssessioncachingredispolling


Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source

Related Questions