23 articles tagged Performance.

Parquet row groups only skip data when the layout cooperates. Why unsorted writes make min/max statistics useless, and how to fix it.
The TweakPad Team

Firefox 144 completed the set at 90% support. How the snapshot model works, why one CSS rule animates a whole page, and the traps.
The TweakPad Team

Deployment settles the format argument, not compression. The real support numbers, and the four lines that serve all of them.
The TweakPad Team
INP measures the worst interaction on a page, not the average. Why 200ms is the bar, and the three phases that consume it.
The TweakPad Team
Why a promise resolves before setTimeout(0), how the microtask queue drains between tasks, and the loop that freezes a page.
The TweakPad Team
Single field, compound, multikey, text, wildcard, hashed, geospatial and TTL — what each is for, and the limits.
The TweakPad Team
Lighthouse cannot measure INP. How to collect LCP, INP and CLS from real users, and why attribution is what matters.
The TweakPad Team
CALCULATE inside a calculated column makes it depend on every column in the table. That is why two harmless columns collide.
The TweakPad Team
Minifying CSS saves 6% after Brotli. Minifying JS saves 36%. The difference is one step compression cannot do: mangling.
The TweakPad Team
createIndex syntax, unique and partial and TTL options, building on a live replica set, and how to confirm the index is actually being used.
The TweakPad Team
Equality, Sort, Range — why index field order decides whether MongoDB sorts in memory, and how to prove it with explain.
The TweakPad Team
Random UUID keys fragment your index. UUIDv7 fixes the ordering and keeps client-side generation. When the switch is worth making.
The TweakPad Team
Catastrophic backtracking turns a working pattern into a CPU bomb. Why nested quantifiers explode, and how to spot and rewrite them.
The TweakPad Team
LCP, INP and CLS with their thresholds, the specific cause of each, and how lab data differs from the field data Google ranks on.
The TweakPad Team
Why streams exist, the four types, and how pipeline() fixes the error handling and back-pressure bugs that pipe() leaves you with.
The TweakPad Team
Compound index ordering, the ESR rule, covered queries, and how to read explain() output to confirm an index is actually being used.
The TweakPad Team
How to diagnose a slow query with EXPLAIN ANALYZE, the index rules that matter, and the eight query patterns that defeat indexes.
The TweakPad Team
How next/image prevents layout shift, what sizes and priority really do, and the config mistakes that make images slower.
The TweakPad Team
The Request Memoization, Data Cache, Full Route Cache and Router Cache — what each one caches, how long for, and how to opt out correctly.
The TweakPad Team
A diagnosis-first guide — how to find the real bottleneck with the Profiler, and the five fixes behind most real wins.
The TweakPad Team
Memoisation has a real price. How to tell which of the three legitimate cases you are in, and why most useMemo is waste.
The TweakPad Team
How the event loop schedules work, why promises run before setTimeout, and how to stop long tasks freezing your page.
The TweakPad Team
How async/await really works, the sequential-await mistake that costs 10x, and how to handle errors and parallel work.
The TweakPad Team