Add a new crate for benchmarks

Currently it only contains one very simplistic benchmark:
Setting 2 metrics and submitting a custom ping.

Why an additional crate?
This way we don't add any new (dev) dependencies to the crates that get released.

Cargo.toml

@@ -4,6 +4,7 @@ members = [
   "glean-core",
   "glean-core/ffi",
   "glean-core/preview",
+  "glean-core/benchmark",
 ]
 
 [profile.release]

Makefile

@@ -67,6 +67,13 @@ test-python: build-python ## Run all Python tests
 
 .PHONY: test test-rust test-rust-with-logs test-kotlin test-swift test-ios-sample
 
+# Benchmarks
+
+bench-rust: ## Run Rust benchmarks
+	cargo bench -p benchmark
+
+.PHONY: bench-rust
+
 # Linting
 
 lint: clippy

glean-core/benchmark/Cargo.toml

@@ -0,0 +1,24 @@
+[package]
+name = "benchmark"
+version = "0.1.0"
+authors = ["Jan-Erik Rediger <jrediger@mozilla.com>"]
+description = "Benchmark the glean-core crate"
+readme = "README.md"
+license = "MPL-2.0"
+edition = "2018"
+publish = false
+
+[lib]
+bench = false
+
+[dependencies]
+# No version specified, this crate never gets published
+glean-core = { path = ".." }
+
+[dev-dependencies]
+tempfile = "3.1.0"
+criterion = "0.3"
+
+[[bench]]
+name = "bench_basic"
+harness = false

glean-core/benchmark/README.md

@@ -0,0 +1,38 @@
+# Glean Benchmarks
+
+The `Glean SDK` is a modern approach for a Telemetry library and is part of the [Glean project](https://docs.telemetry.mozilla.org/concepts/glean/glean.html).
+
+## Benchmarks
+
+This crates provides simple benchmarks for Glean, based on the [criterion benchmark framework](https://bheisler.github.io/criterion.rs/book/criterion_rs.html).
+The library itself does not contain additional code.
+
+### Available benchmarks
+
+* [`benches/bench_basic.rs`](benches/bench_basic.rs) - Setting metrics and submitting a custom ping
+
+### How to run the benchmarks
+
+From the top-level directory of the repository run:
+
+```
+cargo bench -p benchmark
+```
+
+This is also available as a `make` task:
+
+```
+make bench-rust
+```
+
+Any additional run compare results to the preceding run.
+
+### Results
+
+After running the benchmarks, an HTML-rendered report can be found in `target/criterion/report/index.html`.
+We do not provide any results here as the absolute numbers are unreliable.
+Benchmark timings across code changes are more important.
+
+### Why an additional crate?
+
+This way we don't add any new (dev) dependencies to the crates that get released.

glean-core/benchmark/benches/bench_basic.rs

@@ -0,0 +1,48 @@
+use benchmark::glean_core::{metrics::*, CommonMetricData, Configuration, Glean};
+use criterion::{criterion_group, criterion_main, Criterion};
+
+/// Set metrics and submit a custom ping.
+///
+/// Glean, the metrics and the custom ping are instantiated before benchmarking the set/submit
+/// functionality.
+pub fn criterion_benchmark(c: &mut Criterion) {
+    let data_dir = tempfile::tempdir().unwrap();
+    let tmpname = data_dir.path().display().to_string();
+    let cfg = Configuration {
+        upload_enabled: true,
+        data_path: tmpname,
+        application_id: "glean.bench".into(),
+        max_events: None,
+        delay_ping_lifetime_io: false,
+    };
+
+    let mut glean = Glean::new(cfg).unwrap();
+
+    let ping = PingType::new("sample", true, false);
+    glean.register_ping_type(&ping);
+
+    let call_counter: CounterMetric = CounterMetric::new(CommonMetricData {
+        name: "calls".into(),
+        category: "local".into(),
+        send_in_pings: vec!["sample".into()],
+        ..Default::default()
+    });
+
+    let string: StringMetric = StringMetric::new(CommonMetricData {
+        name: "string".into(),
+        category: "local".into(),
+        send_in_pings: vec!["sample".into()],
+        ..Default::default()
+    });
+
+    c.bench_function("glean set and submit", |b| {
+        b.iter(|| {
+            call_counter.add(&glean, 1);
+            string.set(&glean, "hello world");
+            glean.submit_ping(&ping).unwrap();
+        })
+    });
+}
+
+criterion_group!(benches, criterion_benchmark);
+criterion_main!(benches);

glean-core/benchmark/src/lib.rs

@@ -0,0 +1,14 @@
+// This Source Code Form is subject to the terms of the Mozilla Public
+// License, v. 2.0. If a copy of the MPL was not distributed with this
+// file, You can obtain one at https://mozilla.org/MPL/2.0/.
+
+//! # Glean benchmarks
+//!
+//! Benchmarking the glean-core crate. Find benchmarks in the `benches/` directory.
+//!
+//! All documentation for Glean can be found online:
+//!
+//! ## [The Glean SDK Book](https://mozilla.github.io/glean)
+
+// Re-export glean_core in total.
+pub use glean_core;