Catalogue

Cards
165
Records
486
Workloads
19
Corpus version
74847648e25a
Workload
Kind
Setup

73 of 486 records match; the first 20 are shown.

KindRecordOrganisationYearSetup
Failure

Daily peak event burst saturated shared database primary, and with no automatic circuit breaker the throttling was manual

github-2026-actions-db-saturation#f1

GitHub2026
Failure

Brief pod reduction during routine rollout saturated remaining sites, OOMed service-mesh sidecars, and triggered invalid-job runner retries

github-2026-actions-rollout-saturation#f1

GitHub2026Multi-region
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-load-balancer-sidecar-overload#f1

GitHub2026Multi-region
Failure

Connection pools that recycled every 15 minutes synchronized into a thundering herd after a managed Postgres upgrade

clerk-2025-database-connection-recycling#f1

Clerk2025
Decision

Split feature engineering into distributed PySpark pre-processing on Databricks and vertical Python transformations on SageMaker

zalando-2025-inventory-optimisation#d1

Zalando2025
Decision

Use LightGBM with Nixtla MLForecast to combine demand forecast training and inference in one SageMaker job

zalando-2025-inventory-optimisation#d2

Zalando2025
Decision

Use dual offline and online SageMaker Feature Store tiers and write interactive updates back to the offline store

zalando-2025-inventory-optimisation#d3

Zalando2025
Decision

Serve interactive partner inventory-setting updates asynchronously through SQS, Lambda and event-stream notifications

zalando-2025-inventory-optimisation#d4

Zalando2025
Decision

Isolate every pipeline run on dedicated transient Databricks job clusters and SageMaker jobs via zFlow

zalando-2025-inventory-optimisation#d5

Zalando2025
Decision

Rebuild fragmented async systems incrementally in phases instead of writing a replacement from scratch

dropbox-2025-async-platform#d1peak of 100,000 requests/s or more

Dropbox2025
Decision

Standardize ingestion, ordering, delayed scheduling, and dynamic flow control into separate MSM layers

dropbox-2025-async-platform#d2peak of 100,000 requests/s or more

Dropbox2025
Decision

Use health-aware push routing and Atlas-backed SOA lambdas for autoscaling and automated release rollbacks

dropbox-2025-async-platform#d3peak of 100,000 requests/s or more

Dropbox2025
Decision

Batch ping writes with COPY in one transaction instead of calling a stored procedure once per ping

healthchecks-2025-ping-ingest#d1peak under 100,000 requests/s

Healthchecks.io2025One database
Decision

Cache not-found results for UUID ping URLs so repeated pings to missing checks skip the database

healthchecks-2025-ping-ingest#d2peak under 100,000 requests/s

Healthchecks.io2025One database
Decision

Replace sequential monolithic RepositoryPushJob with a Kafka topic fanning out to isolated per-owner background jobs

github-2024-push-processing#d1

GitHub2024
Decision

Use consistent per-event feature flagging to cut over push processing without data loss or double processing

github-2024-push-processing#d2

GitHub2024
Failure

Recurring incidents where errors on the Pushes MySQL cluster early in RepositoryPushJob kept unrelated pull requests from synchronizing

github-2024-push-processing#f1

GitHub2024
Decision

Moved jobs from Resque on Redis to a database-backed queue, starting with the most critical ones

37signals-2023-solid-queue#d1

37signals2023
Decision

Database queue design that keeps the polled table small and claims jobs with SKIP LOCKED

37signals-2023-solid-queue#d2

37signals2023
Decision

Split data quality checking into four loosely coupled backend services connected by SQS queues

lyft-2023-verity-data-quality#d1

Lyft2023

53 more records. Sign in to see all 73.