Ticket Booking — System Design Academy

Complete System Design Tutorial

The design separates the latency-sensitive online path, the correctness-critical durable transition, and asynchronous projection or repair work. Each chapter below explains one decision shown in the diagrams.

1. Requirements and scope

Functional requirements

Execute the primary ticket booking workflow and return an observable result.
Expose the concrete operations shown below with stable retry identities.
Allow operators to inspect, disable, repair, or replay ambiguous state.
Keep derived analytics and notifications off the correctness-critical path.

Non-functional requirements

Design for 10M users at onsale · 100K requests/s · zero double-sells.
Protect the named business invariant under concurrency and partial failure.
Scale by an explicit aggregate/partition key without one global mutex.
Degrade intentionally when cache, queue, store, or regional ownership is unavailable.

Primary invariant: no scarce resource or monetary effect is created twice, lost after acknowledgment, or committed without an audit trail.

↑ top

2. Capacity assumptions

State assumptions before choosing technology. The supplied workload is 10M users at onsale · 100K requests/s · zero double-sells. Convert it into peak throughput, retained records, bandwidth, and failure budget; then connect each number to a component.

Dimension	Working estimate	Design implication
Traffic	10M users at onsale · 100K requests/s · zero double-sells	Separate online compute, durable state, and asynchronous consumers.
Peak factor	5–10× average	Admission control, autoscaling headroom, and hot-key protection.
Stored state	write rate × retention × record bytes	Partition count, indexes, compaction, backup, and archival tier.
Network	peak operations × payload size	Batching, compression, edge placement, and egress budget.
Availability	99.9–99.99%	Multi-zone replicas, tested failover, and dependency error budgets.
Recovery	RPO/RTO by invariant	Ack boundary, replay retention, reconciliation, and operator runbooks.

Do not cargo-cult the numbers: show the arithmetic, name the uncertain variable, and say which production metric will replace the interview estimate.

↑ top

3. API and data model

GET /v1/events/{id}/seatsversioned availability

POST /v1/holdsatomic seat hold

POST /v1/ordersidempotent checkout

POST /v1/orders/{id}/refundsrestore according to policy

Authoritative records and indexes

seat(event_id, seat_id, state, version, hold_id)

hold(hold_id PK, buyer_id, seats, expires_at, state)

order(order_id PK, hold_id, payment_intent, state)

ticket(ticket_id, order_id, barcode_hash, status)

Command and event envelopes

# Command envelope
operation: GET /v1/events/{id}/seats
request_id: req_01J...
idempotency_key: idem_7f2...
aggregate_key: ticket-booking_123
expected_version: 17
payload: {...}

# Durable event
event_id: evt_01J...
aggregate_key: ticket-booking_123
event_type: ticket_booking_committed
schema_version: 3
trace_id: tr_8ac...

The authoritative aggregate and idempotency record share a transactional or conditional-write boundary. Event streams, search indexes, caches, analytics, and dashboards are rebuildable projections.

↑ top

4. Why this architecture and what it costs

The boxes are not a shopping list. Each boundary exists because it owns a different invariant, scaling dimension, data lifecycle, security policy, or failure domain. The table makes those choices explicit and shows when the design should remain simpler.

Component	Owned responsibility	Why separate it	Rejected alternative	Cost and evolution trigger
Buyer Client	Owns client protocol, request identity, retry, and resume behavior. In this design it participates in: Admit a bounded cohort from the waiting room and issue a signed queue token.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents the external caller from also owning the contract and failure behavior of Bot Defense.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Two buyers select the same seat” to the external caller.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Buyer Client boundary before extracting or scaling it.
Bot Defense	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Read approximate availability, but create holds with transactional/conditional seat updates.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Buyer Client from also owning the contract and failure behavior of Virtual Waiting Room.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Payment succeeds after hold timeout” to Buyer Client.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Bot Defense boundary before extracting or scaling it.
Virtual Waiting Room	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Schedule hold expiry and make release idempotent against confirmation.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Bot Defense from also owning the contract and failure behavior of Catalog API.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Expiry worker races confirmation” to Bot Defense.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Virtual Waiting Room boundary before extracting or scaling it.
Catalog API	Owns authentication, validation, quotas, deadlines, and routing. In this design it participates in: Authorize payment, confirm the hold, issue tickets, and append the order ledger.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Virtual Waiting Room from also owning the contract and failure behavior of Availability Cache.	Call every downstream consumer synchronously. That couples user latency and availability to analytics, notification, indexing, and audit dependencies. Here that simpler choice would couple “Waiting-room tokens are replayed” to Virtual Waiting Room.	Costs eventual consistency, schema evolution, replay tooling, idempotent consumers, and lag operations. Earn it when bursts or downstream failure must not block acknowledgment. Measure the latency, saturation, lag, or mismatch at the Catalog API boundary before extracting or scaling it.
Availability Cache	Serves a low-latency, disposable projection with authoritative fallback. In this design it participates in: If payment outcome is uncertain, reconcile by intent before releasing or charging again.	Removes repeated work from the latency-critical path. It prevents Catalog API from also owning the contract and failure behavior of Inventory Shard.	Read authoritative storage for every request. Simpler initially, but latency and source-of-truth load scale together and a hot key becomes a database incident. Here that simpler choice would couple “Refund and ticket scan overlap” to Catalog API.	Costs invalidation, staleness, stampede protection, and memory. Add or split it only after the workload (10M users at onsale · 100K requests/s · zero double-sells) proves repeated reads dominate. Measure the latency, saturation, lag, or mismatch at the Availability Cache boundary before extracting or scaling it.
Inventory Shard	Owns authoritative versioned state or immutable evidence. In this design it participates in: Admit a bounded cohort from the waiting room and issue a signed queue token.	Defines the source of truth and commit boundary. It prevents Availability Cache from also owning the contract and failure behavior of Reservation Service.	Let each service persist its own interpretation. That removes one clear commit boundary and makes retries, support, and reconciliation ambiguous. Here that simpler choice would couple “Two buyers select the same seat” to Availability Cache.	Costs schema migration, replication, backup, and shard ownership. Split storage only when access patterns, retention, or compliance boundaries materially differ. Measure the latency, saturation, lag, or mismatch at the Inventory Shard boundary before extracting or scaling it.
Reservation Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Read approximate availability, but create holds with transactional/conditional seat updates.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Inventory Shard from also owning the contract and failure behavior of Hold Expiry Wheel.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Payment succeeds after hold timeout” to Inventory Shard.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Reservation Service boundary before extracting or scaling it.
Hold Expiry Wheel	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Schedule hold expiry and make release idempotent against confirmation.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Reservation Service from also owning the contract and failure behavior of Order Orchestrator.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Expiry worker races confirmation” to Reservation Service.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Hold Expiry Wheel boundary before extracting or scaling it.
Order Orchestrator	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Authorize payment, confirm the hold, issue tickets, and append the order ledger.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Hold Expiry Wheel from also owning the contract and failure behavior of Payment Gateway.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Waiting-room tokens are replayed” to Hold Expiry Wheel.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Order Orchestrator boundary before extracting or scaling it.
Payment Gateway	Owns authentication, validation, quotas, deadlines, and routing. In this design it participates in: If payment outcome is uncertain, reconcile by intent before releasing or charging again.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Order Orchestrator from also owning the contract and failure behavior of Ticket Issuer.	Put protocol, authentication, quotas, and business transitions in one process. Acceptable at low scale, but it couples external traffic risk to domain releases. Here that simpler choice would couple “Refund and ticket scan overlap” to Order Orchestrator.	Costs another hop, policy distribution, and a high-blast-radius fleet. Introduce it when many clients or services need one protocol and protection boundary. Measure the latency, saturation, lag, or mismatch at the Payment Gateway boundary before extracting or scaling it.
Ticket Issuer	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Admit a bounded cohort from the waiting room and issue a signed queue token.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Payment Gateway from also owning the contract and failure behavior of Order Ledger.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Two buyers select the same seat” to Payment Gateway.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Ticket Issuer boundary before extracting or scaling it.
Order Ledger	Owns authentication, validation, quotas, deadlines, and routing. In this design it participates in: Read approximate availability, but create holds with transactional/conditional seat updates.	Defines the source of truth and commit boundary. It prevents Ticket Issuer from also owning the contract and failure behavior of Refund Service.	Let each service persist its own interpretation. That removes one clear commit boundary and makes retries, support, and reconciliation ambiguous. Here that simpler choice would couple “Payment succeeds after hold timeout” to Ticket Issuer.	Costs schema migration, replication, backup, and shard ownership. Split storage only when access patterns, retention, or compliance boundaries materially differ. Measure the latency, saturation, lag, or mismatch at the Order Ledger boundary before extracting or scaling it.
Refund Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Schedule hold expiry and make release idempotent against confirmation.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Order Ledger from also owning the contract and failure behavior of Audit / Fraud.	Keep this behavior inside an adjacent service. Prefer that simpler option until independent scale, ownership, policy, or failure isolation is measurable. Here that simpler choice would couple “Expiry worker races confirmation” to Order Ledger.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Refund Service boundary before extracting or scaling it.
Audit / Fraud	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Authorize payment, confirm the hold, issue tickets, and append the order ledger.	Proves correctness and supports controlled recovery. It prevents Refund Service from also owning the contract and failure behavior of operators and repair tooling.	Infer correctness from request logs. Logs do not prove durable business state, detect silent drift, or safely repair ambiguous outcomes. Here that simpler choice would couple “Waiting-room tokens are replayed” to Refund Service.	Costs an interface, deployment, telemetry, ownership, and failure mode. Keep it in-process first; extract when independent scale or reliability pays for the boundary. Measure the latency, saturation, lag, or mismatch at the Audit / Fraud boundary before extracting or scaling it.

Decision 01

Chosen boundary vs simpler monolith

Represent seat state with versioned conditional writes; never trust a read followed by an ordinary write.

A single deployable with one transactional store is the right starting point for a small ticket booking workload. The proposed boundary becomes justified when represent seat state with versioned conditional writes; never trust a read followed by an ordinary write. Separating it contains the named failure “Two buyers select the same seat” without placing unrelated work on the critical path.

Decision 02

Correctness vs availability

Create expiring holds, process payment idempotently, then confirm or release inventory.

The design coordinates only where no scarce resource or monetary effect is created twice, lost after acknowledgment, or committed without an audit trail. Everywhere else it accepts bounded staleness or asynchronous propagation. The alternative—strong coordination for every projection—simplifies the mental model but increases tail latency and turns partitions into broad write unavailability.

Decision 03

Operational cost and evolution

Use a virtual waiting room and randomized admission to protect the database during flash sales.

This choice adds dashboards, ownership, schema compatibility, capacity planning, and recovery runbooks. Start with the minimum topology that supports “Schedule hold expiry and make release idempotent against confirmation.” and extract another service only when measured throughput, blast radius, data lifecycle, or team ownership changes independently.

Why this design: the synchronous path contains only the decisions required to protect no scarce resource or monetary effect is created twice, lost after acknowledgment, or committed without an audit trail. Durable projections, notifications, indexing, analytics, and repair are isolated so their load or failure cannot corrupt the acknowledged operation.

Correctness boundary

Represent seat state with versioned conditional writes; never trust a read followed by an ordinary write.

Scale boundary

Create expiring holds, process payment idempotently, then confirm or release inventory.

Operational boundary

Use a virtual waiting room and randomized admission to protect the database during flash sales.

↑ top

5. Critical flow and algorithm

Admit a bounded cohort from the waiting room and issue a signed queue token.

Read approximate availability, but create holds with transactional/conditional seat updates.

Schedule hold expiry and make release idempotent against confirmation.

Authorize payment, confirm the hold, issue tickets, and append the order ledger.

If payment outcome is uncertain, reconcile by intent before releasing or charging again.

Retry-safe mutation skeleton

identity = authenticate(request)
prior = idempotency.get(identity, request.key)
if prior: return prior.result

current = authoritative_store.read(partition_key)
next_state, events = domain.transition(current, request)
authoritative_store.compare_and_set(current.version, next_state)
idempotency.commit(request.key, next_state.result)
outbox.append(events)
return next_state.result

Exactly-once business effect: the network can duplicate requests. Stable identities, conditional state transitions, immutable event IDs, and reconciliation make the user-visible effect happen once.

↑ top

6. Caching, hot keys, and backpressure

Cache policy

Cache immutable or slowly changing reads with versioned keys, TTL jitter, short negative entries, and request coalescing.

Hot partition or tenant

Measure per-key and per-tenant skew. Replicate safe reads, isolate the offender, salt only when ordering permits, and enforce budgets.

Backpressure

Bound queues, buffers, connection pools, and concurrency. Shed low-priority work and scale from lag/oldest-item age rather than CPU alone.

Stampede and retry storms

Use single-flight/coalescing, jittered expiry, bounded retries, exponential backoff, and circuit breakers so one dependency failure does not multiply traffic.

Cache outage rule: never allow every miss to fall through simultaneously. Use local protection, request coalescing, circuit breakers, bounded pools, and controlled fallback.

↑ top

7. Partitioning and consistency

partition = stable_hash(primary_or_aggregate_key) → shard owner
write = conditional_update(expected_version)
read_projection = versioned and rebuildable

Strong boundary

no scarce resource or monetary effect is created twice, lost after acknowledgment, or committed without an audit trail. Use a uniqueness constraint, serializable/conditional write, expected version, or fencing token at this boundary.

Eventual boundary

Analytics, notifications, search, read models, and operational dashboards may lag when freshness is observable and replay exists.

Operation	Consistency	Reason
Authoritative mutation	Conditional / serializable	Protect uniqueness, ownership, money, inventory, order, or legal state transition.
Read after own write	Primary, session token, or cache warm	The caller must observe the result it just committed.
Ordinary read projection	Bounded stale where safe	Replicas and caches improve latency when version/freshness is visible.
Analytics and operations	Eventual + replay	Freshness is less important than isolating the online path.

Route every mutation for one aggregate to one owner/partition epoch.
Use cursors rather than offsets for changing ordered collections.
Retain tombstones until replication and repair windows close.
Move shard ownership through versioned maps and dual-read verification.

↑ top

8. Failure drills and recovery

Failure	Runtime behavior	Recovery proof
Two buyers select the same seat	Resolve at source of truth with uniqueness, expected version, lease generation, or monotonic cursor.	Reconciliation returns zero invariant gaps; lag/replica health and user-visible SLO return to baseline.
Payment succeeds after hold timeout	Query/retry with the same operation identity; do not issue a second business effect.	Reconciliation returns zero invariant gaps; lag/replica health and user-visible SLO return to baseline.
Expiry worker races confirmation	Resolve at source of truth with uniqueness, expected version, lease generation, or monotonic cursor.	Reconciliation returns zero invariant gaps; lag/replica health and user-visible SLO return to baseline.
Waiting-room tokens are replayed	Stop unsafe progress, serve only explicitly safe degraded behavior, and recover from durable state.	Reconciliation returns zero invariant gaps; lag/replica health and user-visible SLO return to baseline.
Refund and ticket scan overlap	Resolve at source of truth with uniqueness, expected version, lease generation, or monotonic cursor.	Reconciliation returns zero invariant gaps; lag/replica health and user-visible SLO return to baseline.

Recovery checkpoint

# Recovery checkpoint
scenario: Two buyers select the same seat
operation_id: op_01J...
aggregate_key: ticket-booking_123
last_durable_version: 18
projection_checkpoint: partition-07@offset-482901
retry_policy: same_identity / bounded_backoff
reconciliation_status: pending
owner: Audit / Fraud

Every dependency needs a timeout, retry budget, circuit policy, and explicit fail-open, fail-closed, stale, or read-only mode. “Highly available” is incomplete until degraded behavior is named.

↑ top

9. Security, regions, and observability

Security and abuse

Authenticate at ingress and authorize the exact resource/tenant on every operation.
Encrypt transport and durable data; tokenize or isolate sensitive material.
Rate-limit by tenant, actor, resource, and network; detect enumeration and anomalous fan-out.
Redact secrets and personal data before logs or analytical storage.
Audit privileged mutations, support access, policy changes, and forced recovery.

Signals that prove correctness

User-visible success rate and p50/p95/p99 latency.
Timeouts, circuit state, saturation, and replica health by dependency.
Queue lag, oldest-item age, retries, and dead-letter growth.
Conflict, duplicate-suppression, and reconciliation mismatch counts.
Business drill: Two buyers select the same seat — alert before the user-visible invariant is violated.

Regional evolution

Begin multi-zone with tested failover, backups, and reconciliation.
Add global stateless ingress and regional read projections.
Keep one write owner per aggregate or partition where coordination is expensive.
Move ownership with epochs/fencing; avoid silent simultaneous writers.
Use expand/contract schemas, canaries, shadow reads, and reversible cutovers.

↑ top

10. Interviewer follow-ups

How do adjacent-seat requests work?

Represent seat state with versioned conditional writes; never trust a read followed by an ordinary write. Walk the concrete lifecycle step “Admit a bounded cohort from the waiting room and issue a signed queue token.” Then test it against “Two buyers select the same seat”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

What if payment succeeds after a hold expires?

Create expiring holds, process payment idempotently, then confirm or release inventory. Walk the concrete lifecycle step “Read approximate availability, but create holds with transactional/conditional seat updates.” Then test it against “Payment succeeds after hold timeout”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do general-admission tickets differ?

Use a virtual waiting room and randomized admission to protect the database during flash sales. Walk the concrete lifecycle step “Schedule hold expiry and make release idempotent against confirmation.” Then test it against “Expiry worker races confirmation”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do you stop bots?

Represent seat state with versioned conditional writes; never trust a read followed by an ordinary write. Walk the concrete lifecycle step “Authorize payment, confirm the hold, issue tickets, and append the order ledger.” Then test it against “Waiting-room tokens are replayed”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do refunds restore inventory?

Create expiring holds, process payment idempotently, then confirm or release inventory. Walk the concrete lifecycle step “If payment outcome is uncertain, reconcile by intent before releasing or charging again.” Then test it against “Refund and ticket scan overlap”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

Answer pattern: state the invariant, choose a mechanism, walk one failure, name the recovery signal, and explain the cost of the alternative.

↑ top

11. How to present this in 45 minutes

Time	What to cover
0–5 min	Clarify scope, workload, SLOs, and the correctness invariant.
5–10 min	Estimate throughput/storage and define concrete APIs and records.
10–20 min	Draw online, durable-write, and asynchronous/repair paths.
20–30 min	Explain the exact lifecycle, partition key, and concurrency control.
30–38 min	Discuss caching, hot keys, replication, and backpressure.
38–43 min	Run failure drills, security, observability, and regional evolution.
43–45 min	Summarize the five most important choices and trade-offs.

Final decision summary

Invariant	no scarce resource or monetary effect is created twice, lost after acknowledgment, or committed without an audit trail.
Partitioning	Stable aggregate/key hash with versioned ownership and measured hot-key isolation.
Commit	Expected version or uniqueness at the authoritative state boundary.
Async work	Durable outbox/log, idempotent consumers, checkpoints, and replay.
Recovery	Same operation identity, durable truth lookup, and business reconciliation.

Closing summary

CLOSING “The design protects no scarce resource or monetary effect is created twice, lost after acknowledgment, or committed without an audit trail. The online path stays bounded, authoritative state is versioned and partitioned, and asynchronous work is replayable and observable.”

↑ top

Buyer Client

Bot Defense

Virtual Waiting Room

Catalog API

Complete System Design Tutorial

1. Requirements and scope

Functional requirements

Non-functional requirements

2. Capacity assumptions

3. API and data model

Authoritative records and indexes

Command and event envelopes

4. Why this architecture and what it costs

Chosen boundary vs simpler monolith

Correctness vs availability

Operational cost and evolution

5. Critical flow and algorithm

Retry-safe mutation skeleton

6. Caching, hot keys, and backpressure

7. Partitioning and consistency

Strong boundary

Eventual boundary

8. Failure drills and recovery

Recovery checkpoint

9. Security, regions, and observability

Security and abuse

Signals that prove correctness

Regional evolution

10. Interviewer follow-ups

11. How to present this in 45 minutes

Final decision summary

Closing summary