Files
Oxicloud/examples/bench_people_list.rs
Claude aba89c4f5d perf: eliminate N+1 hot-path queries, cache immutable lookups, stop re-compressing compressed bytes
Every change is benchmark-verified (harness + before/after numbers in
benches/, measured on this branch; reproduction commands in each doc):

DAV / sync-client hot paths
- PROPFIND dead-properties: one = ANY($1) query per 500-child page instead
  of one sequential query per child, and indexable `=` predicates instead
  of IS NOT DISTINCT FROM (seq scans). 2,000-child folder: 1.07-4.54 s of
  DB chatter -> 4-6 ms (258-773x). Applied to native + NC PROPFIND and
  both NC REPORT handlers. [benches/DEAD-PROPS.md]
- Folder paging: keyset cursor (name > $last) + new partial index
  (folder_id, name) replaces LIMIT/OFFSET full-folder rescan per page.
  Full 20k-file walk: 1266 ms -> 77 ms (16.5x). New migration
  20260917000000. [benches/PROPFIND-PAGING.md]
- NC chroot / default-drive resolution: moka caches (30 s TTL, explicit
  invalidation on drive mutations) for find_default_for_user and the
  markerless chroot FolderDto. 2 uncached queries + 2 pool checkouts per
  NC/WebDAV/WOPI request -> sub-us moka hit (p50 0.7-3.6 ms -> ~1 us).
  [benches/CHROOT-CACHE.md]
- Quota: PROPFINDs whose prop list never names a quota prop skip the
  2-query resolution entirely (wants_quota()); the remaining lookups read
  2 columns instead of the full auth.users row with its <=512 KiB avatar
  (11-16x, p50 3.4 ms -> 0.29 ms). Same narrow read now gates every
  upload quota check. [benches/QUOTA-PATH.md]

CPU on the request path
- ZIP exports (folder download, share ZIP, batch download): entries whose
  MIME says already-compressed (JPEG/MP4/zip/pdf/...) are Stored instead
  of Deflate - deflate ran inline on the tokio writer task at ~41 MB/s
  for ~0% size gain. Mixed media corpus: 4.31x wall and CPU, archive size
  unchanged. Shared predicate in common::mime_detect. [benches/ZIP-MEDIA.md]
- Compression layers: tower-http's default maps to Brotli QUALITY 11
  (verified in brotli-8.0.2 source and empirically: 90 ms per 64 KiB JSON
  response, 1.3 s per 700 KiB bundle). Both layers pinned to Precise(4):
  99x less CPU for ~15% more bytes. SPA assets are now precompressed at
  build time (scripts/precompress.mjs, 77% smaller) and served via
  ServeDir::precompressed_br/gzip: 2016x less per-request work, and
  clients get the better q11 bytes. [benches/STATIC-PRECOMPRESSED.md]

Batched / cached backend paths [benches/NPLUS1-AND-CACHES.md]
- Content-search ReBAC re-verification: new
  AuthorizationEngine::check_files_read_batch (default = old loop;
  PgAclEngine override batches drive resolution + reuses role cache).
  200 sequential point SELECTs per search -> 1-2 queries.
- Batch-ZIP subtree downloads: drop per-file re-authz + per-file Recent
  recording (2 writes/file) for subtree entries already authorized at the
  root - mirrors the native folder-download path. ~6,000 statements
  removed from a 2,000-file archive.
- CDC chunk manifests: immutable by content address, now moka-cached
  (weight-bounded 32 MiB, 60 s TTL, positive-only, invalidated on delete)
  - removes one manifest query (p50 0.44-4.4 ms) from every stream,
  range and full blob read.
- People tab: grouped COUNT + batched cover lookup instead of dragging
  every face row with its 2 KiB embedding (10k faces: 30.4 ms & 21 MB ->
  3.8 ms & 1.3 KB, 8.1x); merge() is one set-based UPDATE.
  [benches/PEOPLE-LIST.md]
- Photos timeline cursor: raw timestamptz comparison instead of
  EXTRACT(EPOCH ...) wrapper + IS NULL OR disjunction - cursor is an
  index boundary again, deep scroll stops re-scanning skipped rows.
- Public share landing: one atomic UPDATE ... access_count + 1 (was
  SELECT + full-row write-back: racy, lost updates, clobbered concurrent
  owner edits) - 3 round-trips -> 2 per visit.
- move_to_trash: dead full-entity SELECT feeding a documented no-op
  removed from both branches; dead fields dropped from TrashService.
- NFC normalization: is_nfc_quick fast path skips the decompose/recompose
  state machine for the ~100% already-NFC case (every row loaded from PG).

Frontend
- Large folders paint after page one (~200 items) via fetchFolderListing's
  new onPage hook instead of waiting for every sequential page.
- Tested-and-reverted (kept for the record): cached Intl.Collator for name
  sorts - vitest showed it 2x SLOWER than V8's argument-less localeCompare
  fast path (5.6 ms vs 12.1 ms / 5k names). Sort order untouched.

New bench harnesses under examples/ (bench feature): zip_media,
dead_props, chroot_cache, quota_path, people_list, propfind_paging,
static_precompress.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CBK1RdtzyP6759Muqe1K1w
2026-07-16 14:20:20 +00:00

214 lines
7.2 KiB
Rust
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
//! People-tab benchmark — full faces scan (embeddings included) vs grouped COUNT.
//!
//! `PeopleService::list_people` used to call `faces_for_user`, dragging every
//! face row — each with a 2,048-byte embedding BYTEA — across the wire and
//! decoding it into a fresh `Vec<f32>`, only to (a) count faces per person and
//! (b) resolve ~a-handful of cover faces to file ids. The change replaces it
//! with `person_face_stats` (grouped COUNT) + `file_ids_for_faces` (one
//! `= ANY` over just the cover ids).
//!
//! Run (needs Postgres up; reads DATABASE_URL from .env):
//! cargo run --release --features bench --example bench_people_list
//! Tunables: BENCH_FACES (10000), BENCH_PERSONS (20), BENCH_REPS (5)
use std::env;
use std::time::Instant;
use sqlx::PgPool;
use sqlx::postgres::PgPoolOptions;
use uuid::Uuid;
fn env_or<T: std::str::FromStr>(key: &str, default: T) -> T {
env::var(key)
.ok()
.and_then(|v| v.parse().ok())
.unwrap_or(default)
}
struct Seeded {
user_id: Uuid,
drive_id: Uuid,
cover_ids: Vec<Uuid>,
}
async fn seed(pool: &PgPool, faces: usize, persons: usize) -> Seeded {
let mut tx = pool.begin().await.expect("begin");
let user_id: Uuid = sqlx::query_scalar(
"INSERT INTO auth.users (username, email, role)
VALUES ('bench_people', 'bench_people@bench.invalid', 'user') RETURNING id",
)
.fetch_one(&mut *tx)
.await
.expect("user");
let drive_id: Uuid = sqlx::query_scalar(
"INSERT INTO storage.drives (kind, default_for_user) VALUES ('personal', $1) RETURNING id",
)
.bind(user_id)
.fetch_one(&mut *tx)
.await
.expect("drive");
let folder_id: Uuid = sqlx::query_scalar(
"INSERT INTO storage.folders (name, path, lpath, drive_id)
VALUES ('bench_people', '/bench_people', 'bench_people', $1) RETURNING id",
)
.bind(drive_id)
.fetch_one(&mut *tx)
.await
.expect("folder");
sqlx::query("UPDATE storage.drives SET root_folder_id = $1 WHERE id = $2")
.bind(folder_id)
.bind(drive_id)
.execute(&mut *tx)
.await
.expect("stamp root");
tx.commit().await.expect("commit");
// Photo files the faces point at.
let file_ids: Vec<Uuid> = sqlx::query_scalar(
"INSERT INTO storage.files (name, folder_id, blob_hash, size, mime_type, drive_id)
SELECT 'p' || i, $1, 'benchpeople0000000000000000000000000000000000000000000000000000',
1024, 'image/jpeg', $2
FROM generate_series(1, $3) AS i
RETURNING id",
)
.bind(folder_id)
.bind(drive_id)
.bind(faces as i32)
.fetch_all(pool)
.await
.expect("files");
// Persons + faces (2 KiB embedding each, like the real 512×f32).
let mut person_ids = Vec::with_capacity(persons);
for i in 0..persons {
let pid: Uuid = sqlx::query_scalar(
"INSERT INTO faces.persons (user_id, display_name) VALUES ($1, $2) RETURNING id",
)
.bind(user_id)
.bind(format!("Person {i}"))
.fetch_one(pool)
.await
.expect("person");
person_ids.push(pid);
}
let embedding = vec![0u8; 2048];
let mut cover_ids = Vec::with_capacity(persons);
for (i, file_id) in file_ids.iter().enumerate() {
let pid = person_ids[i % persons];
let face_id: Uuid = sqlx::query_scalar(
"INSERT INTO faces.faces
(file_id, user_id, person_id, bbox, det_score, quality, embedding, blob_hash)
VALUES ($1, $2, $3, ARRAY[0.1,0.1,0.2,0.2]::real[], 0.99, 0.9, $4,
'benchpeople0000000000000000000000000000000000000000000000000000')
RETURNING id",
)
.bind(file_id)
.bind(user_id)
.bind(pid)
.bind(&embedding)
.fetch_one(pool)
.await
.expect("face");
if i < persons {
cover_ids.push(face_id);
}
}
sqlx::query("ANALYZE faces.faces").execute(pool).await.ok();
Seeded {
user_id,
drive_id,
cover_ids,
}
}
async fn cleanup(pool: &PgPool, s: &Seeded) {
let _ = sqlx::query("DELETE FROM storage.drives WHERE id = $1")
.bind(s.drive_id)
.execute(pool)
.await;
let _ = sqlx::query("DELETE FROM auth.users WHERE id = $1")
.bind(s.user_id)
.execute(pool)
.await;
}
fn median(mut xs: Vec<f64>) -> f64 {
xs.sort_by(|a, b| a.partial_cmp(b).unwrap());
xs[xs.len() / 2]
}
#[tokio::main(flavor = "multi_thread")]
async fn main() {
dotenvy::dotenv().ok();
let url = env::var("DATABASE_URL").expect("set DATABASE_URL");
let faces: usize = env_or("BENCH_FACES", 10_000);
let persons: usize = env_or("BENCH_PERSONS", 20);
let reps: usize = env_or("BENCH_REPS", 5);
let pool = PgPoolOptions::new()
.max_connections(5)
.connect(&url)
.await
.expect("connect");
println!("seeding {faces} faces / {persons} persons (one-time)…");
let seeded = seed(&pool, faces, persons).await;
println!(
"\n# GET /api/people data fetch: BEFORE (full face rows) vs AFTER (COUNT + cover ANY)"
);
println!("{:<28} {:>12} {:>14}", "mode", "total ms", "bytes moved");
let mut base = None;
for mode in ["BEFORE full-rows", "AFTER count+covers"] {
let mut times = Vec::with_capacity(reps);
let mut bytes = 0usize;
for _ in 0..reps {
let t = Instant::now();
if mode.starts_with("BEFORE") {
// faces_for_user shape: every column incl. embedding.
let rows: Vec<(Uuid, Uuid, Option<Uuid>, Vec<u8>)> = sqlx::query_as(
"SELECT id, file_id, person_id, embedding FROM faces.faces WHERE user_id = $1",
)
.bind(seeded.user_id)
.fetch_all(&pool)
.await
.expect("full rows");
bytes = rows.iter().map(|r| r.3.len() + 48).sum();
assert_eq!(rows.len(), faces);
} else {
let stats: Vec<(Uuid, i64)> = sqlx::query_as(
"SELECT person_id, COUNT(*) FROM faces.faces
WHERE user_id = $1 AND person_id IS NOT NULL GROUP BY person_id",
)
.bind(seeded.user_id)
.fetch_all(&pool)
.await
.expect("stats");
let covers: Vec<(Uuid, Uuid)> = sqlx::query_as(
"SELECT id, file_id FROM faces.faces WHERE user_id = $1 AND id = ANY($2)",
)
.bind(seeded.user_id)
.bind(&seeded.cover_ids)
.fetch_all(&pool)
.await
.expect("covers");
bytes = (stats.len() + covers.len()) * 32;
assert_eq!(stats.len(), persons);
}
times.push(t.elapsed().as_secs_f64() * 1000.0);
}
let ms = median(times);
let speedup = base
.map(|b: f64| format!("({:.1}x)", b / ms))
.unwrap_or_default();
println!("{mode:<28} {ms:>12.2} {bytes:>14} {speedup}");
if base.is_none() {
base = Some(ms);
}
}
cleanup(&pool, &seeded).await;
}