Ride-Hailing Service — 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 ride-hailing service 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 5M active drivers · updates every 3–5 s · match < 2 s.
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: each accepted ride-hailing service operation has one stable result and retries do not repeat its business effect.

↑ top

2. Capacity assumptions

State assumptions before choosing technology. The supplied workload is 5M active drivers · updates every 3–5 s · match < 2 s. Convert it into peak throughput, retained records, bandwidth, and failure budget; then connect each number to a component.

Dimension	Working estimate	Design implication
Traffic	5M active drivers · updates every 3–5 s · match < 2 s	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

POST /v1/rides/quotesprice and ETA estimate

POST /v1/ridesidempotent ride request

POST /v1/drivers/{id}/locationsbatched updates

POST /v1/trips/{id}/transitionsversioned state move

Authoritative records and indexes

driver_location(driver_id, cell, lat, lon, heading, observed_at)

trip(trip_id PK, rider_id, driver_id, state, version)

offer(offer_id, trip_id, driver_id, lease_until, state)

pricing_quote(quote_id, inputs_hash, amount, expires_at)

Command and event envelopes

# Command envelope
operation: POST /v1/rides/quotes
request_id: req_01J...
idempotency_key: idem_7f2...
aggregate_key: ride-hailing_123
expected_version: 17
payload: {...}

# Durable event
event_id: evt_01J...
aggregate_key: ride-hailing_123
event_type: ride_hailing_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
Rider App	Owns client protocol, request identity, retry, and resume behavior. In this design it participates in: Map-match driver updates and upsert only newer observations into the geo index.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents the external caller from also owning the contract and failure behavior of Driver App.	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 matchers offer the same driver” 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 Rider App boundary before extracting or scaling it.
Driver App	Owns client protocol, request identity, retry, and resume behavior. In this design it participates in: Search the pickup cell, expand rings, filter eligibility, and rank by ETA.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Rider App from also owning the contract and failure behavior of Realtime 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 “Stale GPS makes nearest driver wrong” to Rider App.	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 Driver App boundary before extracting or scaling it.
Realtime Gateway	Owns authentication, validation, quotas, deadlines, and routing. In this design it participates in: Create short candidate offers; reserve one driver using a conditional lease.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Driver App from also owning the contract and failure behavior of Location Ingest.	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 “Accepted driver disconnects before pickup” to Driver App.	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 Realtime Gateway boundary before extracting or scaling it.
Location Ingest	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Advance trip state with expected version and append each transition to the event trail.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Realtime Gateway from also owning the contract and failure behavior of Map Matcher.	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 “Surge quote expires during request” to Realtime 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 Location Ingest boundary before extracting or scaling it.
Map Matcher	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Use location/trip events for pricing, safety, receipts, and post-trip settlement.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Location Ingest from also owning the contract and failure behavior of Geo Cell Index.	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 “Trip transition arrives out of order” to Location Ingest.	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 Map Matcher boundary before extracting or scaling it.
Geo Cell Index	Serves a low-latency, disposable projection with authoritative fallback. In this design it participates in: Map-match driver updates and upsert only newer observations into the geo index.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Map Matcher from also owning the contract and failure behavior of Supply Service.	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 “Two matchers offer the same driver” to Map Matcher.	Costs invalidation, staleness, stampede protection, and memory. Add or split it only after the workload (5M active drivers · updates every 3–5 s · match < 2 s) proves repeated reads dominate. Measure the latency, saturation, lag, or mismatch at the Geo Cell Index boundary before extracting or scaling it.
Supply Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Search the pickup cell, expand rings, filter eligibility, and rank by ETA.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Geo Cell Index from also owning the contract and failure behavior of Matching Engine.	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 “Stale GPS makes nearest driver wrong” to Geo Cell Index.	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 Supply Service boundary before extracting or scaling it.
Matching Engine	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Create short candidate offers; reserve one driver using a conditional lease.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Supply Service from also owning the contract and failure behavior of Driver Lease Store.	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 “Accepted driver disconnects before pickup” to Supply 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 Matching Engine boundary before extracting or scaling it.
Driver Lease Store	Owns client protocol, request identity, retry, and resume behavior. In this design it participates in: Advance trip state with expected version and append each transition to the event trail.	Defines the source of truth and commit boundary. It prevents Matching Engine from also owning the contract and failure behavior of Trip State 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 “Surge quote expires during request” to Matching Engine.	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 Driver Lease Store boundary before extracting or scaling it.
Trip State Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Use location/trip events for pricing, safety, receipts, and post-trip settlement.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Driver Lease Store from also owning the contract and failure behavior of Pricing Service.	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 “Trip transition arrives out of order” to Driver Lease Store.	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 Trip State Service boundary before extracting or scaling it.
Pricing Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Map-match driver updates and upsert only newer observations into the geo index.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Trip State Service from also owning the contract and failure behavior of ETA / Route Service.	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 matchers offer the same driver” to Trip State 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 Pricing Service boundary before extracting or scaling it.
ETA / Route Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Search the pickup cell, expand rings, filter eligibility, and rank by ETA.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Pricing Service from also owning the contract and failure behavior of Payment Events.	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 “Stale GPS makes nearest driver wrong” to Pricing 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 ETA / Route Service boundary before extracting or scaling it.
Payment Events	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Create short candidate offers; reserve one driver using a conditional lease.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents ETA / Route Service from also owning the contract and failure behavior of Safety / Audit.	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 “Accepted driver disconnects before pickup” to ETA / Route 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 Payment Events boundary before extracting or scaling it.
Safety / Audit	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Advance trip state with expected version and append each transition to the event trail.	Proves correctness and supports controlled recovery. It prevents Payment Events 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 “Surge quote expires during request” to Payment Events.	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 Safety / Audit boundary before extracting or scaling it.

Decision 01

Chosen boundary vs simpler monolith

Keep live locations in an in-memory geo index; durable trip history belongs elsewhere.

A single deployable with one transactional store is the right starting point for a small ride-hailing service workload. The proposed boundary becomes justified when keep live locations in an in-memory geo index; durable trip history belongs elsewhere. Separating it contains the named failure “Two matchers offer the same driver” without placing unrelated work on the critical path.

Decision 02

Correctness vs availability

Expand search cells progressively, score candidates, and use short reservation leases to prevent double assignment.

The design coordinates only where each accepted ride-hailing service operation has one stable result and retries do not repeat its business effect. 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

Model trips as a state machine with idempotent transitions and an append-only event trail.

This choice adds dashboards, ownership, schema compatibility, capacity planning, and recovery runbooks. Start with the minimum topology that supports “Create short candidate offers; reserve one driver using a conditional lease.” 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 each accepted ride-hailing service operation has one stable result and retries do not repeat its business effect. Durable projections, notifications, indexing, analytics, and repair are isolated so their load or failure cannot corrupt the acknowledged operation.

Correctness boundary

Keep live locations in an in-memory geo index; durable trip history belongs elsewhere.

Scale boundary

Expand search cells progressively, score candidates, and use short reservation leases to prevent double assignment.

Operational boundary

Model trips as a state machine with idempotent transitions and an append-only event trail.

↑ top

5. Critical flow and algorithm

Map-match driver updates and upsert only newer observations into the geo index.

Search the pickup cell, expand rings, filter eligibility, and rank by ETA.

Create short candidate offers; reserve one driver using a conditional lease.

Advance trip state with expected version and append each transition to the event trail.

Use location/trip events for pricing, safety, receipts, and post-trip settlement.

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

each accepted ride-hailing service operation has one stable result and retries do not repeat its business effect. 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 matchers offer the same driver	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.
Stale GPS makes nearest driver wrong	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.
Accepted driver disconnects before pickup	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.
Surge quote expires during request	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.
Trip transition arrives out of order	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.

Recovery checkpoint

# Recovery checkpoint
scenario: Two matchers offer the same driver
operation_id: op_01J...
aggregate_key: ride-hailing_123
last_durable_version: 18
projection_checkpoint: partition-07@offset-482901
retry_policy: same_identity / bounded_backoff
reconciliation_status: pending
owner: Safety / Audit

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 matchers offer the same driver — 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 you prevent two riders getting one driver?

Keep live locations in an in-memory geo index; durable trip history belongs elsewhere. Walk the concrete lifecycle step “Map-match driver updates and upsert only newer observations into the geo index.” Then test it against “Two matchers offer the same driver”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How does surge pricing work?

Expand search cells progressively, score candidates, and use short reservation leases to prevent double assignment. Walk the concrete lifecycle step “Search the pickup cell, expand rings, filter eligibility, and rank by ETA.” Then test it against “Stale GPS makes nearest driver wrong”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do you handle GPS noise?

Model trips as a state machine with idempotent transitions and an append-only event trail. Walk the concrete lifecycle step “Create short candidate offers; reserve one driver using a conditional lease.” Then test it against “Accepted driver disconnects before pickup”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do drivers reconnect?

Keep live locations in an in-memory geo index; durable trip history belongs elsewhere. Walk the concrete lifecycle step “Advance trip state with expected version and append each transition to the event trail.” Then test it against “Surge quote expires during request”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How would multi-region matching work?

Expand search cells progressively, score candidates, and use short reservation leases to prevent double assignment. Walk the concrete lifecycle step “Use location/trip events for pricing, safety, receipts, and post-trip settlement.” Then test it against “Trip transition arrives out of order”. 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	each accepted ride-hailing service operation has one stable result and retries do not repeat its business effect.
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 each accepted ride-hailing service operation has one stable result and retries do not repeat its business effect. The online path stays bounded, authoritative state is versioned and partitioned, and asynchronous work is replayable and observable.”

↑ top

Rider App

Driver App

Realtime Gateway

Location Ingest

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