System design

Turn a product requirement into a workload model, measurable objectives and a justified initial capacity estimate.

1. SD1.1Functional scope
2. SD1.2Quality requirements
3. SD1.3Workload estimates
4. SD1.4Capacity and headroom
5. SD1.5Data and bandwidth
6. SD1.6Constraints and decisions

Trace a client request and design an HTTP contract with clear failure, retry and pagination behavior.

1. SD2.1DNS and connection setup
2. SD2.2HTTP semantics
3. SD2.3Resource contracts
4. SD2.4Pagination
5. SD2.5Timeouts and retries
6. SD2.6Idempotency keys

Choose representations and storage boundaries from access patterns, ownership and retention requirements.

1. SD3.1Entities and invariants
2. SD3.2Relational and document models
3. SD3.3Key-value and access patterns
4. SD3.4Files, blocks and objects
5. SD3.5Normalization and read models
6. SD3.6Retention and deletion

Explain how indexes change work and how transactions protect application invariants under concurrency.

1. SD4.1Index access
2. SD4.2Composite indexes
3. SD4.3Write and storage costs
4. SD4.4Transactions and atomicity
5. SD4.5Isolation and anomalies
6. SD4.6Optimistic concurrency

Design a cache key, freshness policy and miss path while calculating load and mean-latency effects.

1. SD5.1Cache identity
2. SD5.2Cache-aside
3. SD5.3Freshness and invalidation
4. SD5.4Eviction and admission
5. SD5.5CDNs and HTTP
6. SD5.6Stampedes and failures

Route requests across instances and calculate capacity with headroom, startup time and instance failures.

1. SD6.1Horizontal and vertical growth
2. SD6.2Routing policies
3. SD6.3Health and readiness
4. SD6.4Application state
5. SD6.5Autoscaling delays
6. SD6.6Failure capacity

Compare consistency contracts and distribute data while tracing stale reads, partitions and hot keys.

1. SD7.1Replication paths
2. SD7.2Consistency models
3. SD7.3Network partitions
4. SD7.4Quorum overlap
5. SD7.5Sharding
6. SD7.6Rebalancing and conflict handling

Design asynchronous processing with explicit acknowledgment, ordering, replay and duplicate-handling rules.

1. SD8.1Queues and logs
2. SD8.2Acknowledgments
3. SD8.3Idempotent consumers
4. SD8.4Ordering and partitions
5. SD8.5Outbox and sagas
6. SD8.6Backpressure and poison jobs

Explain consensus, logical ordering, leases and fencing under stated failure and timing assumptions.

1. SD9.1Failure detection
2. SD9.2Logical ordering
3. SD9.3Consensus and replicated logs
4. SD9.4Quorums and progress
5. SD9.5Leases and fencing
6. SD9.6Transactions and coordination

Define service objectives and use telemetry, failure handling and recovery procedures to protect user outcomes.

1. SD10.1Indicators and objectives
2. SD10.2Error budgets
3. SD10.3Metrics, logs and traces
4. SD10.4Timeouts, backoff and isolation
5. SD10.5Recovery and backups
6. SD10.6Incidents and learning

Place authentication, object authorization and resource limits at explicit application boundaries.

1. SD11.1Authentication and authorization
2. SD11.2Tenant boundaries
3. SD11.3Transport and secrets
4. SD11.4Validation and resource use
5. SD11.5Rate limits
6. SD11.6Auditing and abuse handling

Produce an end-to-end design whose data paths, capacities and failure behavior satisfy explicit requirements.

1. SD12.1A URL shortener
2. SD12.2A chat service
3. SD12.3A feed
4. SD12.4A file-processing pipeline
5. SD12.5Coursefy progress
6. SD12.6Design review