Benchmark the HTTP hosts against axum

A C load generator, an axum service with the same document contract (in
memory, fdatasync, and crash-safe replacement), and a runner that pins the
servers and the load to separate CPUs. On the Pi 5, Pure uses about 40%
less memory, wins with a new connection per request and with eight durable
writers (3.2x the crash-safe axum), but loses 5x on keep-alive reads and
1.5x on 64 durable writers: no persistent connections, one thread, eight
connection slots and two fsyncs per commit. Note 075 records the numbers
and the targets.

1 files changed, 17 insertions, 0 deletions

diff --git a/benchmarks/http/axum-baseline/Cargo.toml b/benchmarks/http/axum-baseline/Cargo.toml
new file mode 100644
index 0000000..1a26451
--- /dev/null
+++ b/benchmarks/http/axum-baseline/Cargo.toml
@@ -0,0 +1,17 @@
+[package]
+name = "axum-baseline"
+version = "0.1.0"
+edition = "2021"
+publish = false
+
+# A reference service for Pure's HTTP benchmarks: the same document contract
+# as Pure's document server, written the way an axum service usually is.
+[dependencies]
+axum = "0.8"
+tokio = { version = "1", features = ["rt-multi-thread", "macros", "net", "fs", "sync", "signal"] }
+bytes = "1"
+
+[profile.release]
+lto = "fat"
+codegen-units = 1
+panic = "abort"