Back on my "essays as commit messages" shithttps://github.com/rust-lang/crates.io/pull/2203 …
-
Show this thread
-
Replying to @sgrif
Why not use a redis or memcached for counting downloads and writing those batchwise to the database?
1 reply 0 retweets 0 likes -
Replying to @recmor
Because we aren't currently running redis or similar in production, and this isn't something that justifies adding to the stack.
2 replies 0 retweets 1 like -
Replying to @sgrif
In that case, just parse the access logs and update the counts that way
1 reply 0 retweets 0 likes -
Replying to @recmor
That's been investigated and there's a variety of reasons we can't do that. Either way the actual write of each download is not causing any issues. 100% of our performance problems are in the background job which updates various other caches
1 reply 0 retweets 1 like
Changing where the writes happen won't improve that performance. For the access patterns we have, an RBDMS is absolutely what we need. Writes *will* eventually be a bottleneck, but not until we have several orders of magnitude more traffic.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.