mirror of
https://github.com/YGGverse/aquatic.git
synced 2026-03-31 17:55:36 +00:00
Update TODO
This commit is contained in:
parent
5ef0935c97
commit
2aa94d050b
1 changed files with 10 additions and 8 deletions
18
TODO.md
18
TODO.md
|
|
@ -17,17 +17,19 @@
|
||||||
* cargo-deny
|
* cargo-deny
|
||||||
|
|
||||||
* aquatic_udp
|
* aquatic_udp
|
||||||
* shard torrent state
|
* notes
|
||||||
* old note that might be useful:
|
* load testing shows that with sharded state, mio reaches 1.4M responses per second
|
||||||
* consider adding ConnectedScrapeRequest::Scrape(PendingScrapeRequest)
|
with 6 socket and 4 request workers. performance is great overall and faster than
|
||||||
containing TransactionId and BTreeMap<usize, InfoHash>, and same for
|
without sharding. io_uring impl is slightly behind or slighly ahead of mio, but
|
||||||
response
|
nothing justifying code complexity and unsafety
|
||||||
|
* clean torrent map in workers, remove it from shared state
|
||||||
|
* consider rewriting load test to just have one worker type. Connection state
|
||||||
|
should/could be divided by socket worker anyway?
|
||||||
|
* mio
|
||||||
|
* stagger connection cleaning intervals?
|
||||||
* uring
|
* uring
|
||||||
* ValidUntil periodic update
|
* ValidUntil periodic update
|
||||||
* statistics
|
* statistics
|
||||||
* shared config keys such as poll interval
|
|
||||||
* mio
|
|
||||||
* stagger connection cleaning intervals?
|
|
||||||
|
|
||||||
* aquatic_http:
|
* aquatic_http:
|
||||||
* clean out connections regularly
|
* clean out connections regularly
|
||||||
|
|
|
||||||
Loading…
Add table
Add a link
Reference in a new issue