Catalogue
- Cards
- 165
- Records
- 486
- Workloads
- 19
- Corpus version
- 74847648e25a
94 of 486 records match; the first 20 are shown.
| Kind | Record | Organisation | Year | Setup |
|---|---|---|---|---|
| Failure | Traffic peak overloaded one datacenter's load balancers while a service-mesh sidecar failed to scale, and a client retry bug amplified authentication traffic | GitHub | 2026 | Multi-region |
| Failure | Connection pools that recycled every 15 minutes synchronized into a thundering herd after a managed Postgres upgrade | Clerk | 2025 | |
| Failure | A dormant topology error and non-staged legacy rollout withdrew the BGP routes of Cloudflare's 1.1.1.1 Resolver globally, making it unavailable for 62 minutes | Cloudflare | 2025 | Multi-region |
| Failure | Automatic removal of IP bindings on 23 percent of edge servers delayed full recovery after BGP readvertisement | Cloudflare | 2025 | Multi-region |
| Decision | Replace hard-coded prefix-to-datacenter lists with service topologies that support staged rollouts | Cloudflare | 2025 | Multi-region |
| Failure | Search compaction invalidations triggered a WebSocket reconnect storm that was prolonged by a manual deploy onto default free-tier hardware | Convex | 2025 | |
| Decision | Build a two-level global and per-location container scheduler on Workers, Durable Objects, and KV | Cloudflare | 2024 | Multi-region |
| Decision | Keep the container scheduler runtime-agnostic across gVisor, Firecracker microVMs, and QEMU | Cloudflare | 2024 | Multi-region |
| Decision | Serve large GPU images via Workers serverless-registry on R2 with a custom zstd and 500 MB chunking CLI | Cloudflare | 2024 | Multi-region |
| Decision | Build an eBPF and Unimog Global State Router sidecar to keep single-IP anycast routing for selectively placed containers | Cloudflare | 2024 | Multi-region |
| Decision | Schedule Workers Builds on nighttime off-peak capacity using prewarmed single-use Firecracker VMs | Cloudflare | 2024 | Multi-region |
| Decision | Scale Pushy exponentially on connection count and raise per-node target from 60,000 to 200,000 connections | Netflix | 2024 | |
| Decision | Rewrite the fixed-size Mantis message processor as a standalone Spring Boot service | Netflix | 2024 | |
| Decision | Migrate the Push Registry from Dynomite to Netflix's managed KeyValue service | Netflix | 2024 | |
| Decision | Shift backend callers from the async queue to direct push for synchronous delivery feedback | Netflix | 2024 | |
| Decision | Route device-to-device messages over a generic JSON layer with Kafka presence tracking and local hot-path caching | Netflix | 2024 | |
| Decision | Add Pushy heartbeats, idle cleanup and client reconnect improvements to catch silent disconnects | Netflix | 2024 | |
| Decision | Replace streaming zlib with tuned streaming zstandard on gateway WebSockets across all client platforms | Discord | 2024 | |
| Decision | Dark-launch compression on live gateway traffic before building clients, then roll out behind an experiment | Discord | 2024 | |
| Decision | Stay with static streaming zstandard parameters rather than shipping dictionaries or dynamic off-peak buffer upgrades | Discord | 2024 |
74 more records. Sign in to see all 94.