SRE-739: Fix backend Deploy concurrency — dropped builds and out-of-order mutable tags

[TimDiekmann]

🌟 What is the purpose of this PR?

The backend Deploy workflow was intermittently cancelling Docker builds on main (e.g. Build kratos (arm64) → "The operation was canceled"). Root cause: the shared …-publish concurrency group dropped pending main-push runs (GitHub keeps only one pending run per group). Combined with the per-commit turbo affected change detection and the promote walk-back, a dropped run for a service-changing commit could let promote ship a stale image to production with no error.

This reworks the workflow's concurrency, manifest assembly, mutable-tag handling, and build cache so that no run is dropped and mutable tags only ever move forward.

🔗 Related links

• SRE-739 (internal)

🚫 Blocked by

• none

🔍 What does this change?

• Concurrency: the workflow-level group is now github.run_id (unique per run → never evicted); serialization moved to job-level queue: max per resource. Builds run unserialized (their pushes are immutable), so the slow Rust builds no longer block each other. Only the mutable tags/rollouts serialize — :staging in a dedicated stage job, :latest in manifest, ECS in deploy (which runs needs: stage, so the redeploy pulls the freshly-advanced :staging rather than the previous image).
• by-digest manifests: GHCR images are pushed by digest; the multi-arch manifest is assembled from those exact digests instead of reading back the moving :latest-<arch> tags. Removes the "Frankenstein manifest" race (arm64 from one commit + amd64 from another) and drops the per-arch GHCR tags.
• Mutable-tag guard (tag-if-newer action): :staging / :latest advance onto a commit only if it is a strict git-descendant of the currently-tagged commit (resolved via the org.opencontainers.image.revision label). Closes the "queue: max FIFO not guaranteed" gap and protects a manual :staging rollback against older/parallel runs.
• Build cache → GHCR registry: moved off the GHA cache backend (type=gha,mode=max was 504-ing on the large Rust builds) to a dedicated private …/<service>/cache:<arch> package. Written on push/workflow_dispatch (trusted events) only; PR/merge_group read it but never write it.
• Retry (nick-fields/retry, repo convention): wraps the imagetools create calls (manifest assemble + guard) against transient GHCR 5xx.

Pre-Merge Checklist 🚀

🚢 Has this modified a publishable library?

This PR:

• does not modify any publishable blocks or libraries, or modifications do not need publishing

📜 Does this require a change to the docs?

The changes in this PR:

• are internal and do not require a docs change

🕸️ Does this require a change to the Turbo Graph?

The changes in this PR:

• do not affect the execution graph

⚠️ Known issues

• The is_main path is not exercised by CI before merge. ECR push (:sha/:run/:staging), the :staging guard, the stage job, and the ECS deploy job all require is_main, so they are skipped on pull_request (PUSH=false) and workflow_dispatch (non-main). They run for the first time on the first main push — that run should be watched. The riskiest unverified assumption is imagetools create against ECR for :staging (ECR's image-index media-type quirk); the fallback is aws ecr put-image.
• The out-of-order guard is logically correct but not empirically tested with two overlapping runs of different commits (the :latest guard only ran single-run).
• First main run: existing :staging/:latest images have no revision label yet → the guard advances them once unconditionally (by design).

🐾 Next steps

• Rollback pin: the guard only moves :staging forward, so a manual rollback isn't durable on its own (the next merge overtakes it). Needs a separate pin/freeze mechanism if "freeze staging" is ever required.
• Move :staging tagging into internal-infra via workflow_call (keep the :latest guard in this OSS repo — it can't go closed-source).

🛡 What tests cover this?

• No automated tests (this is CI workflow config). Validated by a workflow_dispatch run on this branch: all 12 builds parallel, GHCR by-digest + registry cache (no 504), 5 multi-arch manifests assembled from digests, :latest guard. A separate throwaway probe confirmed job-level queue: max serializes without dropping runs. The is_main path is covered only by the first main run after merge.

❓ How to test this?

1. gh workflow run deploy.yml --ref <this branch> → exercises the GHCR path (build-parallel, by-digest, cache, manifest, :latest guard). stage/deploy are skipped (non-main).
2. On merge, watch the first main run for: ECR push, the stage job (:staging guard), and the ECS redeploy — these cannot be exercised beforehand.

📹 Demo

n/a — CI / infra change.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
hash	Ready	Preview, Comment	Jun 1, 2026 9:51pm
hashdotdesign-tokens	Ready	Preview, Comment	Jun 1, 2026 9:51pm
petrinaut	Ready	Preview	Jun 1, 2026 9:51pm

[cursor]

PR Summary

High Risk
Changes production CD (ECR :staging, GHCR :latest, ECS deploy ordering and concurrency); several paths are only fully exercised on first main push after merge.

Overview
Fixes Deploy so main runs are no longer dropped by a shared publish concurrency group (which could skip builds for commits that only that run would publish) and so mutable registry tags cannot move backward when CI finishes out of order.

Concurrency is now one group per run (github.run_id) for push/dispatch; builds run in parallel. Serialization moves to per-service jobs with queue: max: new stage for ECR :staging, manifest for GHCR :latest, and deploy for ECS ( deploy waits on stage so rollouts see the updated :staging).

docker-build-push splits validate / ECR tag push / GHCR push-by-digest, moves layer cache to a private GHCR registry ref (read on PRs; write only on trusted events; skipped for fork PRs), stamps org.opencontainers.image.revision, and returns ghcr-digest. deploy.yml stops tagging :staging and per-arch GHCR tags in build; it uploads digests and assembles :sha-<sha> from those digests instead of :latest-<arch> tags.

New tag-if-newer advances :staging / :latest only when the run’s commit is a strict git descendant of the revision on the current tag (from OCI labels), with retries on transient registry errors.

^{Reviewed by Cursor Bugbot for commit 4c62534. Bugbot is set up for automated code reviews on this repo. Configure here.}

[augmentcode]

🤖 Augment PR Summary

Summary: This PR hardens the backend Deploy workflow against concurrency-related dropped builds and mutable-tag races.

Changes:

• Reworks workflow concurrency: run-level uses unique github.run_id, while mutable-tag/rollout jobs serialize per service via queue: max.
• Build job pushes immutable artifacts in parallel (ECR :sha/:run tags; GHCR pushed by digest only).
• Adds a new tag-if-newer composite action to advance mutable tags (:staging, :latest) only when the candidate commit is a strict descendant of the currently-tagged commit.
• Publishes multi-arch GHCR manifests from per-arch digests (avoids “Frankenstein” manifests from moving per-arch tags).
• Moves build cache from GHA cache backend to a per-service/arch GHCR registry cache, written only on trusted events.
• Adds retry wrappers around docker buildx imagetools create operations to mitigate transient registry failures.

Technical Notes: Images are labeled with org.opencontainers.image.revision to allow tag guards to resolve the commit behind existing tags.

_{🤖 Was this summary useful? React with 👍 or 👎}

[augmentcode]

Review completed. 2 suggestions posted.

Comment augment review to trigger a new review at any time.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 59.07%. Comparing base (41f5267) to head (4c62534).
⚠️ Report is 7 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #8801   +/-   ##
=======================================
  Coverage   59.07%   59.07%           
=======================================
  Files        1344     1344           
  Lines      129750   129750           
  Branches     5868     5868           
=======================================
  Hits        76653    76653           
  Misses      52193    52193           
  Partials      904      904           

Flag	Coverage Δ
apps.hash-ai-worker-ts	1.41% <ø> (ø)
apps.hash-api	0.00% <ø> (ø)
blockprotocol.type-system	40.84% <ø> (ø)
local.claude-hooks	0.00% <ø> (ø)
local.harpc-client	51.24% <ø> (ø)
local.hash-backend-utils	2.81% <ø> (ø)
local.hash-graph-sdk	9.63% <ø> (ø)
local.hash-isomorphic-utils	0.00% <ø> (ø)
rust.antsi	0.00% <ø> (ø)
rust.error-stack	90.87% <ø> (ø)
rust.harpc-codec	84.70% <ø> (ø)
rust.harpc-net	96.19% <ø> (ø)
rust.harpc-tower	67.03% <ø> (ø)
rust.harpc-types	0.00% <ø> (ø)
rust.harpc-wire-protocol	92.23% <ø> (ø)
rust.hash-codec	72.76% <ø> (ø)
rust.hash-graph-api	2.52% <ø> (ø)
rust.hash-graph-authorization	62.34% <ø> (ø)
rust.hash-graph-postgres-store	26.74% <ø> (ø)
rust.hash-graph-store	37.76% <ø> (ø)
rust.hash-graph-temporal-versioning	47.95% <ø> (ø)
rust.hash-graph-types	0.00% <ø> (ø)
rust.hash-graph-validation	83.45% <ø> (ø)
rust.hashql-ast	87.23% <ø> (ø)
rust.hashql-compiletest	28.26% <ø> (ø)
rust.hashql-core	79.28% <ø> (ø)
rust.hashql-diagnostics	72.53% <ø> (ø)
rust.hashql-eval	75.69% <ø> (ø)
rust.hashql-hir	89.06% <ø> (ø)
rust.hashql-mir	86.84% <ø> (ø)
rust.hashql-syntax-jexpr	94.06% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[codspeed-hq]

Merging this PR will not alter performance

✅ 80 untouched benchmarks

---

_{Comparing t/sre-739-fix-backend-deploy-concurrency-dropped-builds-and-out-of (4c62534) with main (9eab148)}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 16728da. Configure here.}

[Copilot]

Pull request overview

Reworks the backend Deploy GitHub Actions workflow to prevent dropped publishing runs and races around mutable Docker tags by switching to per-run concurrency at the workflow level and per-service serialization at the job level, assembling GHCR multi-arch manifests from pushed digests, and adding a “move mutable tag only if newer” guard.

Changes:

• Update deploy workflow concurrency and job structure (new stage job for ECR :staging, per-service serialization for mutable tags and ECS redeploys).
• Push GHCR images by digest and assemble :sha-<sha> manifests from collected digests; advance :latest via a guard action.
• Move build cache to GHCR registry cache refs and introduce a new tag-if-newer composite action.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
.github/workflows/deploy.yml	Changes concurrency model; adds digest-based manifest assembly, a staging tag job, and guarded mutable-tag advancement.
.github/actions/tag-if-newer/action.yml	New composite action to advance mutable tags only when the candidate commit is a strict descendant of the currently-tagged commit.
.github/actions/docker-build-push/action.yml	Updates build/push behavior for GHCR digest pushes and GHCR registry-backed caching; adds digest output for downstream manifest assembly.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

Benchmark results

@rust/hash-graph-benches – Integrations

policy_resolution_large

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 2002	$$25.6 \mathrm{ms} \pm 183 \mathrm{μs}\left({\color{gray}0.963 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$3.38 \mathrm{ms} \pm 14.0 \mathrm{μs}\left({\color{gray}-3.344 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1001	$$12.6 \mathrm{ms} \pm 91.1 \mathrm{μs}\left({\color{red}5.53 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 3314	$$43.7 \mathrm{ms} \pm 410 \mathrm{μs}\left({\color{gray}0.804 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$14.7 \mathrm{ms} \pm 173 \mathrm{μs}\left({\color{gray}-2.337 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 1526	$$24.1 \mathrm{ms} \pm 138 \mathrm{μs}\left({\color{gray}2.53 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 2078	$$27.1 \mathrm{ms} \pm 175 \mathrm{μs}\left({\color{gray}3.17 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.76 \mathrm{ms} \pm 20.0 \mathrm{μs}\left({\color{gray}-0.558 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 1033	$$13.6 \mathrm{ms} \pm 116 \mathrm{μs}\left({\color{red}5.00 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_medium

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 102	$$3.72 \mathrm{ms} \pm 20.0 \mathrm{μs}\left({\color{gray}1.14 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.97 \mathrm{ms} \pm 14.3 \mathrm{μs}\left({\color{gray}-1.173 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 51	$$3.33 \mathrm{ms} \pm 20.1 \mathrm{μs}\left({\color{gray}1.06 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 269	$$5.09 \mathrm{ms} \pm 36.0 \mathrm{μs}\left({\color{gray}1.22 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$3.55 \mathrm{ms} \pm 29.4 \mathrm{μs}\left({\color{gray}1.69 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 107	$$4.08 \mathrm{ms} \pm 27.7 \mathrm{μs}\left({\color{gray}1.67 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 133	$$4.34 \mathrm{ms} \pm 31.7 \mathrm{μs}\left({\color{gray}-0.130 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.48 \mathrm{ms} \pm 17.6 \mathrm{μs}\left({\color{gray}1.16 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 63	$$4.06 \mathrm{ms} \pm 30.1 \mathrm{μs}\left({\color{gray}1.68 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_none

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 2	$$2.59 \mathrm{ms} \pm 15.7 \mathrm{μs}\left({\color{gray}-2.215 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.52 \mathrm{ms} \pm 14.6 \mathrm{μs}\left({\color{gray}-0.098 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1	$$2.57 \mathrm{ms} \pm 16.3 \mathrm{μs}\left({\color{gray}-2.592 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 8	$$2.85 \mathrm{ms} \pm 14.3 \mathrm{μs}\left({\color{gray}-3.680 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.67 \mathrm{ms} \pm 13.1 \mathrm{μs}\left({\color{gray}-3.018 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 3	$$2.88 \mathrm{ms} \pm 20.1 \mathrm{μs}\left({\color{gray}-1.635 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_small

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 52	$$3.02 \mathrm{ms} \pm 15.3 \mathrm{μs}\left({\color{gray}0.818 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.88 \mathrm{ms} \pm 43.9 \mathrm{μs}\left({\color{red}8.11 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 25	$$2.98 \mathrm{ms} \pm 19.3 \mathrm{μs}\left({\color{gray}3.42 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 94	$$3.39 \mathrm{ms} \pm 18.6 \mathrm{μs}\left({\color{gray}1.82 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$2.99 \mathrm{ms} \pm 16.2 \mathrm{μs}\left({\color{gray}0.403 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 26	$$3.32 \mathrm{ms} \pm 22.5 \mathrm{μs}\left({\color{gray}1.18 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 66	$$3.33 \mathrm{ms} \pm 18.0 \mathrm{μs}\left({\color{gray}1.83 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.96 \mathrm{ms} \pm 14.6 \mathrm{μs}\left({\color{gray}-0.092 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 29	$$3.29 \mathrm{ms} \pm 17.6 \mathrm{μs}\left({\color{gray}0.061 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_complete

Function	Value	Mean	Flame graphs
entity_by_id;one_depth	1 entities	$$54.8 \mathrm{ms} \pm 348 \mathrm{μs}\left({\color{gray}3.72 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	10 entities	$$45.6 \mathrm{ms} \pm 212 \mathrm{μs}\left({\color{gray}0.601 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	25 entities	$$50.1 \mathrm{ms} \pm 255 \mathrm{μs}\left({\color{gray}-0.862 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	5 entities	$$44.4 \mathrm{ms} \pm 234 \mathrm{μs}\left({\color{gray}0.636 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	50 entities	$$61.8 \mathrm{ms} \pm 403 \mathrm{μs}\left({\color{gray}0.736 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	1 entities	$$62.6 \mathrm{ms} \pm 537 \mathrm{μs}\left({\color{gray}2.30 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	10 entities	$$55.6 \mathrm{ms} \pm 264 \mathrm{μs}\left({\color{gray}1.13 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	25 entities	$$103 \mathrm{ms} \pm 607 \mathrm{μs}\left({\color{gray}0.741 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	5 entities	$$45.8 \mathrm{ms} \pm 258 \mathrm{μs}\left({\color{gray}0.103 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	50 entities	$$289 \mathrm{ms} \pm 988 \mathrm{μs}\left({\color{gray}-0.346 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	1 entities	$$18.9 \mathrm{ms} \pm 111 \mathrm{μs}\left({\color{gray}-2.341 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	10 entities	$$19.9 \mathrm{ms} \pm 95.6 \mathrm{μs}\left({\color{gray}-2.205 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	25 entities	$$20.3 \mathrm{ms} \pm 133 \mathrm{μs}\left({\color{gray}1.91 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	5 entities	$$19.0 \mathrm{ms} \pm 118 \mathrm{μs}\left({\color{gray}-1.715 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	50 entities	$$25.5 \mathrm{ms} \pm 147 \mathrm{μs}\left({\color{gray}2.58 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_linkless

Function	Value	Mean	Flame graphs
entity_by_id	1 entities	$$19.2 \mathrm{ms} \pm 146 \mathrm{μs}\left({\color{gray}-0.168 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10 entities	$$19.4 \mathrm{ms} \pm 88.4 \mathrm{μs}\left({\color{gray}2.26 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	100 entities	$$19.2 \mathrm{ms} \pm 108 \mathrm{μs}\left({\color{gray}-0.455 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	1000 entities	$$20.1 \mathrm{ms} \pm 122 \mathrm{μs}\left({\color{gray}0.688 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10000 entities	$$26.7 \mathrm{ms} \pm 173 \mathrm{μs}\left({\color{gray}0.496 \mathrm{\%}}\right) $$	Flame Graph

representative_read_entity

Function	Value	Mean	Flame graphs
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/block/v/1	$$33.8 \mathrm{ms} \pm 302 \mathrm{μs}\left({\color{gray}0.235 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/book/v/1	$$34.3 \mathrm{ms} \pm 274 \mathrm{μs}\left({\color{gray}2.20 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/building/v/1	$$34.3 \mathrm{ms} \pm 264 \mathrm{μs}\left({\color{gray}-1.582 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/organization/v/1	$$33.9 \mathrm{ms} \pm 329 \mathrm{μs}\left({\color{gray}0.162 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/page/v/2	$$34.1 \mathrm{ms} \pm 293 \mathrm{μs}\left({\color{gray}-0.168 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/person/v/1	$$34.2 \mathrm{ms} \pm 293 \mathrm{μs}\left({\color{gray}1.60 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/playlist/v/1	$$34.1 \mathrm{ms} \pm 268 \mathrm{μs}\left({\color{gray}2.41 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/song/v/1	$$35.2 \mathrm{ms} \pm 322 \mathrm{μs}\left({\color{gray}0.404 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/uk-address/v/1	$$34.6 \mathrm{ms} \pm 282 \mathrm{μs}\left({\color{gray}2.34 \mathrm{\%}}\right) $$	Flame Graph

representative_read_entity_type

Function	Value	Mean	Flame graphs
get_entity_type_by_id	Account ID: bf5a9ef5-dc3b-43cf-a291-6210c0321eba	$$8.44 \mathrm{ms} \pm 35.3 \mathrm{μs}\left({\color{gray}-0.491 \mathrm{\%}}\right) $$	Flame Graph

representative_read_multiple_entities

Function	Value	Mean	Flame graphs
entity_by_property	traversal_paths=0	0	$$90.6 \mathrm{ms} \pm 423 \mathrm{μs}\left({\color{gray}0.020 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$144 \mathrm{ms} \pm 623 \mathrm{μs}\left({\color{gray}1.18 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$97.7 \mathrm{ms} \pm 556 \mathrm{μs}\left({\color{gray}-0.766 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$110 \mathrm{ms} \pm 601 \mathrm{μs}\left({\color{gray}1.88 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$116 \mathrm{ms} \pm 725 \mathrm{μs}\left({\color{gray}1.09 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$124 \mathrm{ms} \pm 564 \mathrm{μs}\left({\color{gray}1.86 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=0	0	$$105 \mathrm{ms} \pm 610 \mathrm{μs}\left({\color{gray}2.78 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$133 \mathrm{ms} \pm 525 \mathrm{μs}\left({\color{gray}2.26 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$111 \mathrm{ms} \pm 638 \mathrm{μs}\left({\color{gray}2.09 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$121 \mathrm{ms} \pm 673 \mathrm{μs}\left({\color{gray}2.69 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$124 \mathrm{ms} \pm 795 \mathrm{μs}\left({\color{gray}4.03 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$123 \mathrm{ms} \pm 493 \mathrm{μs}\left({\color{gray}1.78 \mathrm{\%}}\right) $$

scenarios

Function	Value	Mean	Flame graphs
full_test	query-limited	$$191 \mathrm{ms} \pm 807 \mathrm{μs}\left({\color{gray}0.259 \mathrm{\%}}\right) $$	Flame Graph
full_test	query-unlimited	$$152 \mathrm{ms} \pm 828 \mathrm{μs}\left({\color{lightgreen}-18.217 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-limited	$$41.3 \mathrm{ms} \pm 210 \mathrm{μs}\left({\color{gray}3.89 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-unlimited	$$553 \mathrm{ms} \pm 1.09 \mathrm{ms}\left({\color{gray}-2.147 \mathrm{\%}}\right) $$	Flame Graph