/

Play the game Tutorials How-to guides Explanation Reference Design About

Reference

Research tree, full screen Stripe Billing DNS round robin Load balancing Health checks Managed identity Caching Connection pooling Read replicas Eager loading Query analysis Structured logs Metrics Percentiles Alerting Funnel analytics PHP-FPM OPcache Ruby Node.js Python Node.js cluster mode libuv thread pool Laravel Rails Express Django Symfony Sinatra Flask Connection reuse Compression Timeouts and retries Load balancing algorithms Continuous delivery Blue-green deploys Canary deploys Code review Automated tests Full-text search Object storage Cache strategies Storage lifecycle rules Retention policies Content delivery Web application firewall API gateway WebSockets Background jobs Message queues Stampede protection Cache warming Go Java C#Gin Spring Boot ASP.NET Core Distributed tracing Trace search Notebooks Certificate automation Elixir Rust Zig Haskell Phoenix Checkout Tiered caching Multiple zones Autoscaling Container orchestration Serverless Reserved instances Spot instances SLO dashboards Error budgets Backoff with jitter Circuit breakers Priority load shedding Idempotency keys Read-your-writes routing Go GC tuning Axum Actix Web ZGC Data warehousing Event streaming Embeddings Hosted models Model serving Point-in-time recovery TLS everywhere HTTP/2 Synchronous replication Restore drills Dependency scanning Sharding Encryption at rest Multiple regions GeoDNS Infrastructure as code GitOps Anycast Document databases Edge compute HTTP/3 Notifications Hybrid fan-out Leaderless replication Consensus databases Disaster recovery drills SOC 2 program Data residency controls Security team Investor relations Guidance modeling PHP JIT Node.js worker threads YJIT gevent ASGI with Uvicorn PHP worker mode Swoole Bun Deno Falcon PyPy Java virtual threads Spring WebFlux fasthttp Tokio blocking pool GraalVM native image .NET Native AOT Distributed Erlang NIFs jemalloc Free-threaded Python FastAPI Fastify Next.js NestJS Fiber Quarkus Change data capture Service boundaries Service mesh Feature flags Secrets management Spam filter On-call rotation Runbooks SRE team Enterprise sales Online schema migrations Leap smearing A/B testing Hiring pipeline Trunk-based development Platform team Data team L4 load balancing Rate limiting Edge-side includes Bot management Scheduled scaling ARM instances Quantization Continuous batching Bulkheads Graceful degradation Hedged requests Chaos engineering Denormalization Materialized views Binary RPC between services Transactional outbox Expand and contract Continuous profiling Tail-based sampling Real user monitoring Anomaly detection Credential stuffing protection Penetration testing Rightsizing report Egress optimization Unit economics Cost allocation Savings plans

Runtimes and frameworks

Instances and regions

Exits and Legacy

SpoilersAll Eras show

ReferenceResearch tree

PHP JIT

The PHP 8 JIT: 1.05× CPU on web kinds, 1.6× on CPU-heavy ones. It barely helps page and read.

Prerequisites#

Research these first: OPcache.

PHP JIT

Track: Runtimes
Era: Seed
Cost
Requires: OPcache

See also

Research treeEvery Research item by track and Era: what it costs in Insight, what it needs first, and what it unlocks.

Support Quorum on Ko-fiQuorum is a game and simplifies real systems on purpose. Check the real thing before you decide. It changes every week, so this page can lag behind the game. See the changelog.