Reconciles two independently-developed lines from base 006d3d0:
ours — M9/M10 community layers, retroactive purge + re-materialization,
signal revocation, agent capability boundaries, P1 feedback loop,
reason labels, instrumented metrics
theirs — M11/M12 cluster mode (tidal-net gRPC transport, tidal-server
cluster/scatter-gather, tidal-stress), multi-vector preference,
ANN candidate-gen, warm-tier day buckets, keyed signal snapshots
Notable semantic resolutions:
* storage::keys::Tag — both sides allocated 0x0E..0x11 for different
records. Kept theirs' 0x0E..0x1A (shipped on-disk format) and renumbered
ours to 0x1B..0x1E (CommunityMembership/Revocation/PurgeManifest/
CommunityLeave); Tag::ALL grown to 30 so the contiguity drift guard holds.
* ranking executor — took theirs' rewrite (SignalReadPlan pre-pass, keyed
SignalKey snapshots, Result-returning reads, finalize()) and re-applied
ours' M10 read suppression at the chokepoints it introduced:
single_signal_score, score_hot/trending/controversial,
CreatorEngagementRate, and the Stage-4 boost loop.
* signals::warm — theirs' day-bucket/read-time-rotation rewrite, with ours'
subtract_bucket and Clone extended to the new day tier; ours' test split
kept (warm/tests.rs, warm/proptests.rs) carrying theirs' updated bodies.
* db::signals — kept ours' contribution-logging try_cohort_attribution in
signal_dispatch.rs and theirs' event-time try_update_preference_vector;
dropped the superseded duplicates.
* db::mod / from_parts — theirs' constructors, with ours' purge/
re-materialization/revocation/community/skip-counter fields and restart
rebuilds; from_parts kept in its own file per the 600-line guideline.
* schema::validation::builders — ours' module split with theirs' expanded
tests; policy validation runs both sides' checks (read-signal lists +
profile overrides, then the zero-duration limit guard).
* feedback Unhide no longer writes a -1.0 "hide" signal: theirs' engine
rejects negative weights (spec §8). Reverses index state only, matching
every other undo action.
* SessionState::new is now the single construction path (gains
overrides_rejected/default_profile); AuditEntry gains kind on the
deserialize path, inferred from the accepted flag as before.
* Removed tidal/src/replication/tcp_transport.rs and its test: never
declared in replication/mod.rs on either branch, so it had never
compiled and nothing referenced it. Superseded by tidal-net's
GrpcTransport.
Verified: cargo clippy -p tidaldb (lib) clean; --all-targets compiles for
tidaldb/tidal-net/tidal-server/tidal-stress; 2094/2094 lib tests and the
integration suite pass except m8p3_reconcile_production's two CRDT-count
assertions, which fail identically on MERGE_HEAD (pre-existing).
tidalctl cannot build locally: its aws-sdk deps need rustc 1.91.1, local
toolchain is 1.91.0.
165 lines
4.8 KiB
Rust
165 lines
4.8 KiB
Rust
#![allow(clippy::unwrap_used)]
|
|
//! Benchmarks for the session layer: signal writes, snapshot reads,
|
|
//! and retrieve queries with active session context.
|
|
|
|
use std::{collections::HashMap, time::Duration};
|
|
|
|
use criterion::{Criterion, black_box, criterion_group, criterion_main};
|
|
use tidaldb::{
|
|
TidalDb,
|
|
query::retrieve::{ProfileRef, RetrieveBuilder},
|
|
schema::{AgentPolicy, DecaySpec, EntityId, EntityKind, SchemaBuilder, Timestamp, Window},
|
|
};
|
|
|
|
fn session_schema() -> tidaldb::schema::Schema {
|
|
let mut builder = SchemaBuilder::new();
|
|
let _ = builder
|
|
.signal(
|
|
"reward",
|
|
EntityKind::Item,
|
|
DecaySpec::Exponential {
|
|
half_life: Duration::from_secs(7 * 24 * 3600),
|
|
},
|
|
)
|
|
.windows(&[Window::OneHour, Window::TwentyFourHours])
|
|
.velocity(false)
|
|
.add();
|
|
let _ = builder.session_policy(
|
|
"bench_policy",
|
|
AgentPolicy {
|
|
allowed_signals: vec!["reward".to_string()],
|
|
denied_signals: vec![],
|
|
max_session_duration: Duration::from_secs(3600),
|
|
max_signals_per_session: 1_000_000,
|
|
..AgentPolicy::default()
|
|
},
|
|
);
|
|
builder.build().unwrap()
|
|
}
|
|
|
|
/// Benchmark: `session_signal()` write throughput.
|
|
/// Target: < 200µs per call.
|
|
fn bench_session_signal(c: &mut Criterion) {
|
|
let db = TidalDb::builder()
|
|
.ephemeral()
|
|
.with_schema(session_schema())
|
|
.open()
|
|
.unwrap();
|
|
|
|
let mut meta = HashMap::new();
|
|
meta.insert("title".to_string(), "bench-item".to_string());
|
|
db.write_item_with_metadata(EntityId::new(1), &meta)
|
|
.unwrap();
|
|
|
|
let handle = db
|
|
.start_session(1, "bench-agent", "bench_policy", HashMap::new())
|
|
.unwrap();
|
|
let entity = EntityId::new(1);
|
|
let ts = Timestamp::now();
|
|
|
|
c.bench_function("session_signal", |b| {
|
|
b.iter(|| {
|
|
db.session_signal(
|
|
black_box(&handle),
|
|
black_box("reward"),
|
|
black_box(entity),
|
|
black_box(1.0_f64),
|
|
black_box(ts),
|
|
black_box(None),
|
|
)
|
|
.unwrap();
|
|
});
|
|
});
|
|
}
|
|
|
|
/// Benchmark: `session_snapshot()` on an active session with 100 signals.
|
|
/// Target: < 50µs per call.
|
|
fn bench_session_snapshot(c: &mut Criterion) {
|
|
let db = TidalDb::builder()
|
|
.ephemeral()
|
|
.with_schema(session_schema())
|
|
.open()
|
|
.unwrap();
|
|
|
|
let mut meta = HashMap::new();
|
|
meta.insert("title".to_string(), "bench-item".to_string());
|
|
db.write_item_with_metadata(EntityId::new(1), &meta)
|
|
.unwrap();
|
|
|
|
let handle = db
|
|
.start_session(2, "bench-agent", "bench_policy", HashMap::new())
|
|
.unwrap();
|
|
let session_id = handle.id;
|
|
let entity = EntityId::new(1);
|
|
let ts = Timestamp::now();
|
|
|
|
// Pre-load 100 signals.
|
|
for _ in 0..100 {
|
|
db.session_signal(&handle, "reward", entity, 1.0, ts, None)
|
|
.unwrap();
|
|
}
|
|
|
|
c.bench_function("session_snapshot_100_signals", |b| {
|
|
b.iter(|| db.session_snapshot(black_box(session_id)).unwrap());
|
|
});
|
|
}
|
|
|
|
/// Benchmark: `retrieve()` with FOR SESSION vs without, over 1K items.
|
|
/// Measures the overhead of session context in ranking.
|
|
/// Target: < 5ms overhead vs without session.
|
|
fn bench_retrieve_with_session(c: &mut Criterion) {
|
|
let db = TidalDb::builder()
|
|
.ephemeral()
|
|
.with_schema(session_schema())
|
|
.open()
|
|
.unwrap();
|
|
|
|
// Write 1K items.
|
|
for i in 1u64..=1000 {
|
|
let mut meta = HashMap::new();
|
|
meta.insert("title".to_string(), format!("item-{i}"));
|
|
db.write_item_with_metadata(EntityId::new(i), &meta)
|
|
.unwrap();
|
|
}
|
|
|
|
let handle = db
|
|
.start_session(3, "bench-agent", "bench_policy", HashMap::new())
|
|
.unwrap();
|
|
let session_id = handle.id;
|
|
let ts = Timestamp::now();
|
|
|
|
// Signal 10 entities to create a non-trivial session context.
|
|
for i in 1u64..=10 {
|
|
db.session_signal(&handle, "reward", EntityId::new(i), 1.0, ts, None)
|
|
.unwrap();
|
|
}
|
|
|
|
let query_with = RetrieveBuilder::new(EntityKind::Item, ProfileRef::new("hot"))
|
|
.limit(20)
|
|
.for_session(session_id)
|
|
.build()
|
|
.unwrap();
|
|
|
|
let query_without = RetrieveBuilder::new(EntityKind::Item, ProfileRef::new("hot"))
|
|
.limit(20)
|
|
.build()
|
|
.unwrap();
|
|
|
|
let mut group = c.benchmark_group("retrieve_1k_items");
|
|
group.bench_function("without_session", |b| {
|
|
b.iter(|| db.retrieve(black_box(&query_without)).unwrap());
|
|
});
|
|
group.bench_function("with_session", |b| {
|
|
b.iter(|| db.retrieve(black_box(&query_with)).unwrap());
|
|
});
|
|
group.finish();
|
|
}
|
|
|
|
criterion_group!(
|
|
benches,
|
|
bench_session_signal,
|
|
bench_session_snapshot,
|
|
bench_retrieve_with_session
|
|
);
|
|
criterion_main!(benches);
|