Are there any documents about how QUIC performs?

In April of 2015, we wrote a blog post which discussed the performance benefits we've seen from QUIC.

Can I build the toy client and server without building all of chrome?

Yes, the quic_server target can be built without building the rest of Chrome.  You can follow these instructions to build and run the stand-alone QUIC client and server.

How do I aim Chrome at the test server?

If you have an HTTP server, you'll need it to emit a response header that looks like:

Alternate-Protocol: quic:<QUIC server port>

Then you can just run chrome as usual and it will automatically start using QUIC.

If you're testing only with the toy quic server, you can do something like:

% chrome --disable-setuid-sandbox --enable-quic --origin-to-force-quic-on=localhost:6121 http://localhost:6121/

If you need help troubleshooting, try running the QUIC server with --v=1 or check out playing-with-quic

Why is the transport layer encrypted?

The transport information for QUIC (congestion related information) is encrypted mainly to guarantee the transport can always evolve.  If the acks were in the clear, or even checksummed, the concern is that eventually middle boxes would start parsing the congestion information and would break with any forward changes.  This is currently a problem for TCP; the wire format allows for negotiated options and flexible features which are practically unusable because of the expectations of current hardware on the internet.

The down side, of course, is that hiding these details from middle boxes also means QUIC flows are hard to analyze if you don't control an endpoint.  The tcpdump tool can let one visualize the rate of packets, and the gaps in packets, but it's unclear which packets contain payload, which have congestion information, which have retransmits etc.  Both client and server side code are architected to have hooks to easily dump this information during userspace processing.  Such logs can are more data rich than tcpdumps and can be tied together with kernel level packet traces to get a better idea of latency across the whole system.