apollo_network: broadcast network stress test metrics

[sirandreww-starkware]

No description provided.

[reviewable-StarkWare]

This change is 

[sirandreww-starkware]

• apollo_network: broadcast network stress test metrics #9461 👈 (View in Graphite)
• apollo_network: broadcast_network_stress_test_node updated message #9336
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[github-actions]

Artifacts upload workflows:

• Started at 2025-09-29T13:47:18Z
• Started at 2025-09-30T05:07:32Z
• Started at 2025-10-29T21:00:34Z
• Started at 2025-11-03T08:49:11Z
• Started at 2025-11-09T13:13:15Z
• Started at 2025-11-12T11:29:13Z
• Started at 2025-11-30T07:00:21Z
• Started at 2025-11-30T09:32:16Z

[sirandreww-starkware]

This PR adds metrics that are collected by the stress test.
it add a dependency on sysinfo and thus requires approval from some other team lead, who should I add?

Reviewable status: 0 of 5 files reviewed, all discussions resolved (waiting on @ShahakShama)

[github-actions]

Benchmark movements: No major performance changes detected.

[ShahakShama]

@ShahakShama reviewed 4 of 5 files at r1, 1 of 1 files at r2, all commit messages.
Reviewable status: all files reviewed, 8 unresolved discussions (waiting on @sirandreww-starkware)

---

crates/apollo_network/Cargo.toml line 42 at r2 (raw file):

strum.workspace = true
strum_macros.workspace = true
sysinfo.workspace = true

Add this as bin dependencies
Alternatively, move all the stress test code to a different crate

---

Cargo.toml line 323 at r2 (raw file):

strum_macros = "0.25.2"
syn = "2.0.39"
sysinfo = "0.37.1"

Following the 3rd party discussion, @dan-starkware should approve this

---

crates/apollo_network/src/bin/broadcast_network_stress_test_node/main.rs line 98 at r2 (raw file):

        broadcast_topic_client.broadcast_message(message.clone()).await.unwrap();
        trace!("Sent message {i}: {:?}", message);
        // counter!("sent_messages").increment(1);

Delete instead of commenting out

---

crates/apollo_network/src/bin/broadcast_network_stress_test_node/main.rs line 117 at r2 (raw file):

    let _delay_seconds = duration.as_secs_f64();
    // let delay_micros = duration.as_micros().try_into().unwrap();

Same here

---

crates/apollo_network/src/bin/broadcast_network_stress_test_node/metrics.rs line 27 at r2 (raw file):

define_metrics!(
    Infra => {
        MetricGauge { BROADCAST_MESSAGE_HEARTBEAT_MILLIS, "broadcast_message_theoretical_heartbeat_millis", "The number of ms we sleep between each two consecutive broadcasts" },

Consider adding some word that makes it clear these are the setup/config of the scenario
(If broadcast was supposed to be that word then it's not a good word)

---

crates/apollo_network/src/bin/broadcast_network_stress_test_node/metrics.rs line 83 at r2 (raw file):

    };

    // Use apollo_metrics for all metrics including labeled ones

This comment feels redundant

---

crates/apollo_network/src/bin/broadcast_network_stress_test_node/metrics.rs line 90 at r2 (raw file):

    );

    // Use apollo_metrics histograms for latency measurements

This comment feels redundant

---

crates/apollo_network/src/bin/broadcast_network_stress_test_node/metrics.rs line 155 at r2 (raw file):

        interval.tick().await;

        // Refresh only the specific data we need

Remove commented out code

[dan-starkware]

@dan-starkware reviewed 1 of 5 files at r1.
Reviewable status: all files reviewed, 8 unresolved discussions (waiting on @sirandreww-starkware)