TCP Fast Open: expediting web services | Hacker News Reader

TCP Fast Open: expediting web services

lwn.net

51 pointsdeadhead5 comments

We tried to do this at Orbital Data (WAN optimization firm), and one or more existing TCP implmentations didn't support it. Compatibility with existing implementations was a significant barrier to optimizations that were technically permitted within the spec.

I dont recall exactly which TCPs this broke, but I do remember the implementations for VXWorks, some thirdparty TCP for mainframe OSs (IBM 370), and Solaris were often the culprit, and less often there was a problem with a Windows TCP (remember there are different releases to test of each. If you don't, your customers will). While I realize you may not care about those servers, you really don't want to make changes to the client that are incompatible.

I'd suggest checking with the SPDY guys. They've done tons of testing of various latency improvements.

Seems like it's likely to be very difficult for a browser to detect whether or not a given request is idempotent, no? You could use HTTP verbs (i.e. assume that GETs are and POSTs are not) but then you're trusting the server to use REST correctly.

I had the same sort of thought: it's really, really difficult for the bulk of programmers to create an idempotent operation. Therefore, through no fault of its own, TCP Fast Open would acquire a reputation for goofy behavior.

RFC 2616

> 9.1.2 Idempotent Methods

> The methods GET, HEAD,

> PUT and DELETE share this property. Also, the methods OPTIONS and

> TRACE SHOULD NOT have side effects, and so are inherently idempotent.

RFC 2119

> 4. SHOULD NOT

> This phrase, or the phrase "NOT RECOMMENDED" mean that

> there may exist valid reasons in particular circumstances when the

> particular behavior is acceptable or even useful, but the full

> implications should be understood and the case carefully weighed

> before implementing any behavior described with this label.

What could possibly go wrong?

Obviously we aren't going to be able to blindly enable this optimisation without making sure our apps meet the assumptions first.

What effect this will have on HTTPS (and therefore all HTTP 2.0) handshake latency?

Reply on news.ycombinator.com