Simultaneous upload and download of a 10 MB file took 10.6 s over TCP-encapsulated QUIC, 23.3 s over traditional meek, and 34.9 s over meek with encapsulated QUIC (Table 1), showing that naively adding a QUIC session layer to meek degraded throughput by approximately 50% relative to unmodified meek. Performance was sensitive to HTTP body size limits and request-thread count, but the root cause remained uncertain.
From 2020-fifield-turbo — Turbo Tunnel, a good way to design censorship circumvention protocols
· Table 1, §3.2
· 2020
· Free and Open Communications on the Internet
Implications
When inserting a session/reliability layer into a request-response carrier like meek, aggressively tune maximum HTTP body size and request concurrency — default parameters are likely to degrade performance significantly.
Benchmark session-layer additions against baseline transport performance before deployment; session-layer overhead can exceed the cost of the obfuscation layer itself.