Re: [whatwg/fetch] Concurrent GET requests can cause cache misses? (#932)

7234 does not require that two GET requests be separated by a cache. It is *extremely* common for caches (at least in intermediaries) to "collapse" overlapping requests for the same URL by holding any after the first until the response headers are received; if the response is cacheable, it's fanned out to all of the queued requests.

This usually entails having a marker object in cache for those responses that aren't cacheable, so you don't needlessly block requests for things that probably won't benefit from this. Not sure how important that would be in a browser context, though.

For more info:
  * https://wiki.squid-cache.org/Features/CollapsedForwarding
  * https://cwiki.apache.org/confluence/display/TS/Collapsed+Forwarding
  * https://docs.fastly.com/en/guides/request-collapsing


-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/fetch/issues/932#issuecomment-529741808