Collaborative Document Editor — 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 collaborative document editor 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 100 concurrent editors/doc · < 100 ms local echo · offline edits.
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 collaborative document editor 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 100 concurrent editors/doc · < 100 ms local echo · offline edits. Convert it into peak throughput, retained records, bandwidth, and failure budget; then connect each number to a component.

Dimension	Working estimate	Design implication
Traffic	100 concurrent editors/doc · < 100 ms local echo · offline edits	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

WS APPLY_OP(doc_id, client_seq, base_version, op)

GET /v1/docs/{id}/sync?version=snapshot plus ops

POST /v1/docs/{id}/shareversioned ACL

POST /v1/docs/{id}/exportimmutable revision

Authoritative records and indexes

operation(doc_id, server_seq, actor_id, client_op_id, payload)

snapshot(doc_id, version, content_uri, checksum)

client_cursor(doc_id, device_id, acked_seq)

permission(doc_id, principal, role, version)

Command and event envelopes

# Command envelope
operation: WS APPLY_OP(doc_id, client_seq, base_version, op)
request_id: req_01J...
idempotency_key: idem_7f2...
aggregate_key: collaborative-editor_123
expected_version: 17
payload: {...}

# Durable event
event_id: evt_01J...
aggregate_key: collaborative-editor_123
event_type: collaborative_editor_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
Editor Client	Owns client protocol, request identity, retry, and resume behavior. In this design it participates in: Apply edits locally for instant feedback and enqueue client operation identity.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents the external caller from also owning the contract and failure behavior of Local Document Model.	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 “Offline client returns with compacted base” 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 Editor Client boundary before extracting or scaling it.
Local Document Model	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Gateway routes by document; server rechecks permission and base version.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Editor Client 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 “Permission revoked during active session” to Editor 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 Local Document Model boundary before extracting or scaling it.
Realtime Gateway	Owns authentication, validation, quotas, deadlines, and routing. In this design it participates in: Transform/merge against concurrent operations, assign canonical sequence, and append durably.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Local Document Model from also owning the contract and failure behavior of Session Directory.	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 “Duplicate client operation reconnects” to Local Document Model.	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.
Session Directory	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Broadcast canonical operation; clients reconcile pending local operations and acknowledge cursor.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Realtime Gateway from also owning the contract and failure behavior of Permission 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 operations target deleted range” 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 Session Directory boundary before extracting or scaling it.
Permission Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Snapshot periodically and compact only after active/offline retention guarantees.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Session Directory from also owning the contract and failure behavior of Document Router.	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 “Snapshot checksum differs from op replay” to Session Directory.	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 Permission Service boundary before extracting or scaling it.
Document Router	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Apply edits locally for instant feedback and enqueue client operation identity.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Permission Service from also owning the contract and failure behavior of OT / CRDT 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 “Offline client returns with compacted base” to Permission 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 Document Router boundary before extracting or scaling it.
OT / CRDT Engine	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Gateway routes by document; server rechecks permission and base version.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Document Router from also owning the contract and failure behavior of Canonical Op Log.	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 “Permission revoked during active session” to Document Router.	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 OT / CRDT Engine boundary before extracting or scaling it.
Canonical Op Log	Preserves durable ordering, replay, and consumer checkpoints. In this design it participates in: Transform/merge against concurrent operations, assign canonical sequence, and append durably.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents OT / CRDT Engine from also owning the contract and failure behavior of Snapshot Builder.	Call every downstream consumer synchronously. That couples user latency and availability to analytics, notification, indexing, and audit dependencies. Here that simpler choice would couple “Duplicate client operation reconnects” to OT / CRDT Engine.	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 Canonical Op Log boundary before extracting or scaling it.
Snapshot Builder	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Broadcast canonical operation; clients reconcile pending local operations and acknowledge cursor.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Canonical Op Log from also owning the contract and failure behavior of Document 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 “Two operations target deleted range” to Canonical Op Log.	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 Snapshot Builder boundary before extracting or scaling it.
Document Store	Owns authoritative versioned state or immutable evidence. In this design it participates in: Snapshot periodically and compact only after active/offline retention guarantees.	Defines the source of truth and commit boundary. It prevents Snapshot Builder from also owning the contract and failure behavior of Presence 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 “Snapshot checksum differs from op replay” to Snapshot Builder.	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 Document Store boundary before extracting or scaling it.
Presence Service	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Apply edits locally for instant feedback and enqueue client operation identity.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Document Store from also owning the contract and failure behavior of Cursor Broadcast.	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 “Offline client returns with compacted base” to Document 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 Presence Service boundary before extracting or scaling it.
Cursor Broadcast	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Gateway routes by document; server rechecks permission and base version.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Presence Service from also owning the contract and failure behavior of Offline Merge.	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 “Permission revoked during active session” to Presence 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 Cursor Broadcast boundary before extracting or scaling it.
Offline Merge	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Transform/merge against concurrent operations, assign canonical sequence, and append durably.	Separates a distinct scale, ownership, policy, or failure boundary. It prevents Cursor Broadcast from also owning the contract and failure behavior of Export / 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 “Duplicate client operation reconnects” to Cursor Broadcast.	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 Offline Merge boundary before extracting or scaling it.
Export / Audit	Owns one bounded domain decision, transition, or operational capability. In this design it participates in: Broadcast canonical operation; clients reconcile pending local operations and acknowledge cursor.	Proves correctness and supports controlled recovery. It prevents Offline Merge 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 “Two operations target deleted range” to Offline Merge.	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 Export / Audit boundary before extracting or scaling it.

Decision 01

Chosen boundary vs simpler monolith

Apply local edits optimistically and reconcile operations with OT or a CRDT; define ordering and identity.

A single deployable with one transactional store is the right starting point for a small collaborative document editor workload. The proposed boundary becomes justified when apply local edits optimistically and reconcile operations with ot or a crdt; define ordering and identity. Separating it contains the named failure “Offline client returns with compacted base” without placing unrelated work on the critical path.

Decision 02

Correctness vs availability

Persist an append-only operation log and create periodic snapshots to bound replay time.

The design coordinates only where each accepted collaborative document editor 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

Keep cursors and presence ephemeral; document contents and permissions require durable consistency.

This choice adds dashboards, ownership, schema compatibility, capacity planning, and recovery runbooks. Start with the minimum topology that supports “Transform/merge against concurrent operations, assign canonical sequence, and append durably.” 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 collaborative document editor 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

Apply local edits optimistically and reconcile operations with OT or a CRDT; define ordering and identity.

Scale boundary

Persist an append-only operation log and create periodic snapshots to bound replay time.

Operational boundary

Keep cursors and presence ephemeral; document contents and permissions require durable consistency.

↑ top

5. Critical flow and algorithm

Apply edits locally for instant feedback and enqueue client operation identity.

Gateway routes by document; server rechecks permission and base version.

Transform/merge against concurrent operations, assign canonical sequence, and append durably.

Broadcast canonical operation; clients reconcile pending local operations and acknowledge cursor.

Snapshot periodically and compact only after active/offline retention guarantees.

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

Add a cache only for a measured repeated read; do not place mutable correctness state in a disposable layer.

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 collaborative document editor 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
Offline client returns with compacted base	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.
Permission revoked during active session	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.
Duplicate client operation reconnects	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.
Two operations target deleted range	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.
Snapshot checksum differs from op replay	Isolate the key/tenant/priority, apply admission control, and preserve durable work.	Reconciliation returns zero invariant gaps; lag/replica health and user-visible SLO return to baseline.

Recovery checkpoint

# Recovery checkpoint
scenario: Offline client returns with compacted base
operation_id: op_01J...
aggregate_key: collaborative-editor_123
last_durable_version: 18
projection_checkpoint: partition-07@offset-482901
retry_policy: same_identity / bounded_backoff
reconciliation_status: pending
owner: Export / 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: Offline client returns with compacted base — 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

OT or CRDT—which would you choose?

Apply local edits optimistically and reconcile operations with OT or a CRDT; define ordering and identity. Walk the concrete lifecycle step “Apply edits locally for instant feedback and enqueue client operation identity.” Then test it against “Offline client returns with compacted base”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How does offline editing merge?

Persist an append-only operation log and create periodic snapshots to bound replay time. Walk the concrete lifecycle step “Gateway routes by document; server rechecks permission and base version.” Then test it against “Permission revoked during active session”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do you garbage-collect history?

Keep cursors and presence ephemeral; document contents and permissions require durable consistency. Walk the concrete lifecycle step “Transform/merge against concurrent operations, assign canonical sequence, and append durably.” Then test it against “Duplicate client operation reconnects”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do you undo remote edits?

Apply local edits optimistically and reconcile operations with OT or a CRDT; define ordering and identity. Walk the concrete lifecycle step “Broadcast canonical operation; clients reconcile pending local operations and acknowledge cursor.” Then test it against “Two operations target deleted range”. State the retry identity, durable checkpoint, recovery signal, and cost of the alternative.

How do permissions change mid-session?

Persist an append-only operation log and create periodic snapshots to bound replay time. Walk the concrete lifecycle step “Snapshot periodically and compact only after active/offline retention guarantees.” Then test it against “Snapshot checksum differs from op replay”. 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 collaborative document editor 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 collaborative document editor 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

Editor Client

Local Document Model

Realtime Gateway

Session Directory

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