proxygen

Looking forward to the time when Meta will make https://github.com/facebookincubator/zstrong.git public found it mentioned in https://github.com/facebook/proxygen/blob/main/build/fbcode_..., looks like it's going to be cousin of zstd, but maybe for the stronger compression use cases

Very cool! And congrats for reaching version 1.0.0! Other nice HTTP/3 libraries include H2O: https://h2o.examp1e.net as well as Facebook/Meta ProxyGen: https://github.com/facebook/proxygen

Facebook's Proxygen is also pretty neat: https://github.com/facebook/proxygen

Facebook has also open sourced implementation of HTTP3 (https://github.com/facebook/proxygen#quic-and-http3) and IEFT QUIC (https://github.com/facebookincubator/mvfst).

Fastly: https://github.com/h2o/quicly Facebook: https://github.com/facebook/proxygen/issues/127#issuecomment... Cloudflare: https://blog.cloudflare.com/head-start-with-quic/ Akamai: https://developer.akamai.com/blog/2018/10/10/introducing-qui...

This doesn't really replace something like haproxy. In the diagrams, the "real" servers that they are load balancing to are L7 load balancers like HAProxy (using https://github.com/facebook/proxygen most likely) The problem this solves is is when you are using haproxy but need redundancy or a single server is no longer capable of handling the load and you need to scale out to 2+ servers. You could run 4 haproxy servers and use round robin DNS to load balance between them, but as they say, this has the problem of: > compared to DNS it doesn't need to wait for TTL to redirect traffic from failed L7 lb. Many people likely use anycast/ECMP for this case, but that has the other issue they mention: > compared to anycast-based solution, L4 lbs are more resilient to the networking related issues (which triggers mass ECMP reshuffle)/ enable bette

>nginx run behind it. It removes a lot of complexity from your code. There are good intentions with that advice but I think it's misleading. It doesn't take into account how http libraries like this one (and similar ones for C++ such as Proxygen[1] and Silicon[2]) are supposed to be used. The intended use case is to add an embedded webserver to your executable that's communicating with friendly and internal systems. That would be things like microservices and dashboards. You don't use those libraries to create external public-facing websites that would be under attack from hostile agents. You're correct: Do not re-invent NGINX by using the C http library. As an example proper use case, let's say you write an internal C program to process terabytes of image files. You think it might be nice to have a visual status of its progress/erro

The info in https://lyft.github.io/envoy/docs/intro/comparison.html#prox... about Proxygen not supporting HTTP/2 is not correct. Proxygen has had HTTP/2 support for a while (https://github.com/facebook/proxygen/blob/master/proxygen/li...). Disclaimer: I work on Proxygen at FB.

https://github.com/facebook/proxygen C++, used by Facebook in production. We have been using it in a high performance RTB application and has performed remarkably well.

Not sure if the talk covered it, but the core Proxygen framework is opensource: https://github.com/facebook/proxygen

I'm curious- why C? Strings, scoped objects and C++11 move operators seems much safer and clearer from an API perspective. The complaints about C++ seem to mostly be around the ability to abuse the language, not specific issues that C solves. Something like https://github.com/facebook/proxygen seems like a better API. And I don't quite buy portability- if it's not a modern compiler with decent security checks then I'm note sure it should be building web-facing code.

Facebook's proxygen has HTTP/2 support "in progress": https://github.com/facebook/proxygen

Indeed, Silicon should be seen as a glue between low level http libraries, a message encoding format, others externals resources like databases, and your actual API. You can use it with microhttpd, or any other network libraries (the current version provides backends to cpp-netlib and mimosa https://github.com/abique/mimosa). Future integration with websocket library is planned. Don't know much about proxygen but it seems like the goal of the project are similar: Ease the writing of high performance C++ web APIs. However, when reading proxygen echo sample [1], I would say that silicon has a higher level of abstraction. [1] https://github.com/facebook/proxygen/tree/master/proxygen/ht...

If the author has seen Facebook proxygen[1] (another C++ http framework), a comparison to it would be helpful. At first glance, it looks like the Silicon dependency microhttpd is somewhat analogous to proxygen. Is the Silicon value-added proposition the extra layers on top such as auto gen of client-side code, etc? [1] https://github.com/facebook/proxygen

Hey, I'm a Software Engineer on Proxygen as well. Proxygen heavily relies on folly's buffer management abstractions such as IOBuf (https://github.com/facebook/folly/blob/master/folly/io/IOBuf...) and Cursor (https://github.com/facebook/folly/blob/master/folly/io/Curso...). Protocol parsing implementation uses folly::io::Cursor to safely read byte sequences across non-contiguous buffers. Errors during parsing are wrapped up in Result types (https://github.com/facebook/proxygen/blob/master/proxygen/li...) which take an inspiration from Rust. Such constructs simplify our implementation to a reasonable extent and are still low-level enough to extract performance.