Browser & Frontend Runtime
Compositor layers: promotion, overlap, and GPU memory
You add will-change: transform to one card in a list to make its hover animation smooth. DevTools’ Layer Borders panel lights up with forty coloured boxes you never asked for. The browser promoted your neighbours because it had no choice.
How elements reach the compositor
Some elements get their own compositor layer:
transform: translate3d(...)or any 3D transformwill-change: transform(or any compositable property)position: fixed<video>,<canvas>- Currently-animated
opacity filter: blur(...)and other graphical filters- Isolated stacking contexts that overlap another layer
Each layer is rasterised once (paint), uploaded to the GPU, then re-composited cheaply for the next thousand frames. Animating a layer’s transform is “free” — the main thread can sleep. Animating its top is expensive — the main thread re-layouts the world every frame.
Layer lifecycle
The overlap rule: layers you never asked for
The compositor must maintain visual correctness. If element A is a layer (say, transform) and element B paints on top of A in document order without being a layer, the compositor cannot resolve their stacking — A’s layer would appear on top of B even though B is supposed to be on top.
The browser fixes this by promoting B to its own layer, often dragging adjacent elements with it. The compositor then squashes adjacent non-animating layers into a shared “squashed” layer to limit explosion, but the heuristic is imperfect.
A single innocent will-change: transform on a card in the middle of a feed can promote a dozen siblings.
Use DevTools’ Rendering panel → “Layer Borders” to see what is promoted. Pay attention to layers labelled “compositor-induced” or “overlap” — those crept in without your asking.
will-change as a two-edged sword
will-change: transform is a hint to the browser: “I am about to animate this property, please pre-paint the bitmap so the compositor can move it without going back to layout or paint.” The browser promotes the element immediately.
The cost: GPU memory — width × height × 4 bytes per layer.
The will-change anti-pattern: will-change: transform on every component in a design system permanently reserves a layer per component instance. On a list of 100 cards each containing 5 will-change’d sub-elements, the page holds 500 layers — possibly hundreds of MB of GPU memory. On a phone with 256 MB of GPU memory available, the OS starts evicting; the browser re-rasterises on the fly and the page stutters more than without layers.
The fix: set will-change just before the animation starts (e.g., on mouseenter) and remove it on animationend. Treat it as a fast-forward signal, not a permanent setting.
Edge cases
The overlap rule interacts with z-index stacking contexts in subtle ways. An element with z-index: 1 and position: relative creates an isolated stacking context. If that stacking context overlaps a compositor layer, its entire painted area must be promoted to avoid cross-layer bleed. This is why adding position: relative; z-index: 1 to a tooltip that sits above an animated background can trigger a full-page layer cascade in complex layouts. The solution is to keep promotable elements (animated, fixed, video) in a separate stacking context from non-promotable siblings, or to use isolation: isolate explicitly to contain stacking context effects.
In the kitchen metaphor, a layer is a finished plate sitting on the pass — already cooked, ready for the runner. What is the JavaScript equivalent that turns a normal element into one of these pre-cooked plates so the compositor can move it without re-cooking?
A 1920 × 1080 element is promoted to its own compositor layer. Each pixel is 4 bytes (RGBA). How many megabytes of GPU memory does this single layer consume?
A team adds `will-change: transform` to every card in their design system. GPU memory on mobile balloons. What is the mechanism?
Element A has `will-change: transform`. Element B overlaps A in document order. Element B has no promotion trigger. What does the browser do?
- 01What is the GPU memory cost of a 1920×1080 compositor layer?
- 02Why does `will-change: transform` on one element sometimes promote its neighbours?
- 03What is the correct usage pattern for `will-change`?
A compositor layer is a GPU bitmap that the compositor can position and blend cheaply per frame, without touching the main thread. Promotion triggers include 3D transforms, will-change, position: fixed, video, canvas, and animated opacity. The overlap rule forces the browser to promote elements that overlap a layer — a single will-change in a list can trigger a dozen implicit promotions. Each full-screen layer costs ~7.91 MB of GPU memory; exhausting mobile GPU budgets causes tile eviction and re-rasterisation, producing worse jank than no layers at all. The correct pattern is to scope will-change to the duration of an animation. Layer Borders in DevTools visualises every promoted element and its reason.
appears again in162
- Why GraphQL gets N+1junior
- DataLoader mechanics: tick-boundary batchingmiddle
- Batch function contracts: ordering, shapes, errorsmiddle
- Federation and lookahead: batching beyond DataLoadermiddle
- Query complexity defences: depth, cost, persisted queriesmiddle
- Senior GraphQL API: scheduling contract, tenant isolation, observabilitysenior
- Why idempotency: making retries safejunior
- Server-side state machine: four states of an idempotency keymiddle
- Outbox and inbox: effectively-once across the dual-write boundarymiddle
- Concurrency and cache architecture for idempotency at scalesenior
- Observability, production failures, and global-scale designsenior
- What is a cache stampede and why it makes things worsejunior
- Lock and single-flight: bounding concurrent rebuildsmiddle
- XFetch: coordination-free probabilistic early expirationmiddle
- Stale-while-revalidate and CDN request coalescingmiddle
- Detecting stampedes and designing TTL for productionmiddle
- Metastable failure, fencing tokens, and production postmortemssenior
- What a relation is: tables, rows, keys, and constraintsjunior
- Constraints, keys, and Postgres data typesmiddle
- Normal forms, denormalization, and why schemas stickmiddle
- JSONB, arrays, and when a side table winsmiddle
- Heap storage, TOAST, and column alignmentsenior
- Schema integrity: deferral, versioning, and production failure modessenior
- Relational vs document, wide-column, graph, and key-valuesenior
- Index-only scans, the Visibility Map, and INCLUDEsenior
- Production failure modes and the index audit playbooksenior
- pg_statistic, ANALYZE, and production observabilitymiddle
- Production failure modes and plan stabilitysenior
- MVCC: why readers and writers never wait for each otherjunior
- Row versions and snapshots: the on-disk mechanicsmiddle
- HOT updates and isolation levels: what you gain and what you paymiddle
- Vacuum and bloat: keeping the storage tax boundedmiddle
- CLOG, XID wraparound, and MultiXact: deep visibility internalssenior
- SSI internals and production autovacuum tuningsenior
- Real-world MVCC failures, deployment patterns, and distributed snapshotssenior
- Connection pools: amortising the cost of a Postgres backendjunior
- PgBouncer session, transaction, and statement modesmiddle
- Pool sizing: the (cores × 2) + spindles formula and the two-layer stackmiddle
- Pool exhaustion and idle-in-transaction: the 3 AM failure modemiddle
- Migrating to transaction mode: rollout playbook and PgBouncer 1.21 prepared statementsmiddle
- The Postgres process model and why raising max_connections degrades throughputsenior
- Pooler landscape 2026, serverless connection storms, and the full failure-mode taxonomysenior
- What a schema migration is and why it replaces ad-hoc DDLjunior
- ADD COLUMN: instant in PG 11+ vs rewrite in older Postgresjunior
- The lock-queue failure mode: why instant DDL can freeze the databasemiddle
- Safe DDL patterns: NOT VALID, CONCURRENTLY, and unsafe-op fixesmiddle
- Expand-contract: zero-downtime for breaking schema changesmiddle
- Advisory locks, migration tools, and deploy coordinationsenior
- Migration failure taxonomy and production disciplinesenior
- Why sharding exists: the single-Postgres ceilingjunior
- Shard-key selection: hash, range, list, and directory strategiesmiddle
- Partitioning vs sharding: same word, two different thingsmiddle
- Co-location and Citus: the invariant that makes sharding usablemiddle
- The hot-shard failure mode: detection, isolation, and durable policymiddle
- Schema-based sharding and multi-tenancy alternativessenior
- Online resharding, 2PC, and the operational cost of shardingsenior
- The seven acts: from CREATE TABLE to Citusjunior
- Acts 1–3 in depth: schema, indexes, and planner statisticsmiddle
- Acts 4–6 in depth: MVCC bloat, connection pooling, and safe migrationsmiddle
- Act 7 in depth: sharding, co-location, and the seven-tier tradeoff cascademiddle
- Observability, anti-patterns, and production triagesenior
- Raft roles, terms, and why majority quorums prevent split brainjunior
- How Raft replicates a log entry and decides it is safe to commitmiddle
- Raft leader election: timeouts, voting rules, and the four safety propertiesmiddle
- Raft in the real world: partitions, slow disks, and client routingmiddle
- Raft extensions: pre-vote, learners, snapshots, and linearizable readssenior
- Raft in production: membership changes, Multi-Raft, and observabilitysenior
- Where data fetching happens — and why it decides LCPjunior
- Fetch waterfalls — diagnosis and the Promise.all curemiddle
- React Server Components and Suspense streamingmiddle
- Client-side cache: TanStack Query, SWR, and stale-while-revalidatemiddle
- LCP, prefetch, and race conditions in interactive fetchingmiddle
- Senior internals: RSC payload, caching layers, and production failure modessenior
- The IP envelopejunior
- Reading the IP headermiddle
- The three-way handshakejunior
- Sequence numbers and connection statemiddle
- DNS: what it does and why it existsjunior
- The resolver walk: referrals, record types, and gluemiddle
- TTL, caching, and DNS propagationmiddle
- What TLS does and why it existsjunior
- The 1-RTT handshake: key shares and ECDHEmiddle
- Session resumption and 0-RTTmiddle
- Key schedule, SNI, ALPN, and extensionssenior
- 0-RTT defenses, ECH, hybrid PQ, and production TLSsenior
- WebSocket: the HTTP upgrade handshakejunior
- WebSocket frame format: opcodes, masking, fragmentationmiddle
- WebSocket backpressure: when clients can''''t keep upmiddle
- Reconnection: jittered backoff, thundering herd, message resumptionsenior
- WebSocket at scale: HTTP/2 multiplexing, permessage-deflate, C10Msenior
- WebSocket in production: proxies, security, and distributed architecturesenior
- What reverse proxies dojunior
- Health checks, connection draining, and slow startmiddle
- Session affinity, consistent hashing, and the right fixmiddle
- Retry storms, circuit breakers, and load sheddingsenior
- Resilient LB architecture: anycast, zone-aware routing, and observabilitysenior
- Why QUIC and not TCP+TLSjunior
- Connection IDs and network migrationmiddle
- 0-RTT resumption and packet encryptionsenior
- DDoS: what it is and why it worksjunior
- Amplification attacks and state exhaustionmiddle
- Rate limiting: algorithms and architecturemiddle
- WAFs, firewalls, mTLS, and HSTSmiddle
- DNS cache poisoning and BGP hijackingsenior
- Defense-in-depth architecture and attack economicssenior
- The twelve layers: one URL, seven actorsjunior
- DNS, TCP, TLS in sequence: where the milliseconds gomiddle
- Proxy intercepts and security gates: rate limiters, WAF, mTLSmiddle
- Alternate paths: QUIC 0-RTT, WebSocket upgrade, connection migrationmiddle
- Observability: distributed traces, USE/RED, and samplingsenior
- Resilience: cascading retries, circuit breakers, and error budgetssenior
- What the three signals are: logs, metrics, and tracesjunior
- Why structured logs exist: the diary vs the spreadsheetjunior
- The production log schema: fields every line must carrymiddle
- PII redaction and log injectionsenior
- OTel Logs Data Model and audit logs as a subsystemsenior
- What is OpenTelemetry: API, SDK, Collector, OTLPjunior
- OTel signals, Semantic Conventions, and the OTLP wire formatmiddle
- The OTel Collector: receivers, processors, exporters, and deployment patternsmiddle
- Vendor neutrality, eBPF instrumentation, the Operator, and browser/serverless OTelsenior
- Operating the OTel Collector: reliability, version skew, failure modes, and governancesenior
- SLI, SLO, and the error budget: reliability by the numbersjunior
- Error budget policy, latency SLOs, and composite journeysmiddle
- Production SLO failures, self-observability, security, and the big picturesenior
- What is trace propagation and why broken propagation is worse than nonejunior
- traceparent and tracestate: the W3C header format in fullmiddle
- Baggage and async boundaries: carrying context across queues and callbacksmiddle
- Async context per language, service mesh, B3 migration, and securitysenior
- Production propagation failures, span links, and platform designsenior
- The debugging funnel: SLO → RED → trace → profilejunior
- OTel architecture: one SDK, four signals, one wire formatmiddle
- The incident loop: from pager to postmortem to preventionmiddle
- Scale, security, and the ROI of observable systemssenior
- Cache lines, struct layout, and false sharingmiddle
- SIMD, SoA vs AoS, and memory bandwidthmiddle
- Cache-oblivious algorithms, PGO, and production failuressenior
- GC in production: observability, security, edge cases, and fleet governancesenior
- Batching: amortize fixed cost per operationjunior
- The batching window: size and wait timemiddle
- Batching in Kafka and Postgresmiddle
- io_uring and observability of batchingmiddle
- From Nagle to io_uring: evolution of batchingmiddle
- Backpressure, failure isolation, and batch security in productionsenior
- CI enforcement and RUM: making budgets stickmiddle
- V8 JIT pipeline, HTTP priorities, and bundle securitysenior
- The performance loop: discipline, not a projectjunior
- Classify and fix: matching bottleneck families to remediesmiddle
- Observability stack and CI gates: catching regressions before they shipmiddle
- Incident to enforcement: SLO burn to verified fix in 35 minutesmiddle
- Culture, economics, and org-scale performancesenior
- At-most-once, at-least-once, exactly-once: the three delivery contractsjunior
- The three failure legs — where duplicates and losses actually happenmiddle
- Consumer-side dedup: the cheapest path to exactly-once processingmiddle
- Kafka exactly-once semantics: idempotent producer and transactionsmiddle
- SQS visibility timeout, DLQ, and the outbox patternmiddle
- Exactly-once in production: impossibility proof, hybrid patterns, and real incidentssenior
- What OAuth is and why passwords are not the answerjunior
- Authorization code flow with PKCEmiddle
- ID token validation and JWKS cache managementmiddle
- Refresh token rotation and scope-based least privilegemiddle
- Sender-constrained tokens: DPoP and mTLSsenior
- OAuth in production: audience attacks, observability, and real failuressenior