/orleans

name: orleans
description: "Design Microsoft Orleans systems from each primitive's purpose and failure model. USE FOR: grains, digital twins, state versus databases, transactions, messaging, streams, timers, reminders, Durable Jobs, stateless workers, grain services, startup, and hosting. DO NOT USE FOR: other actor stacks, batches, relational-only CRUD, or advice without an Orleans decision. INVOKES: inspect version and topology, choose the primitive, implement, and validate."

Microsoft Orleans

Start With Purpose

Do not begin with an Orleans API. First state:

1. the business identity that owns the behavior; 2. the invariant and what must survive activation or cluster failure; 3. the required consistency, query shape, acknowledgement, durability, replay, fan-out, and timing.

Then select the smallest Orleans primitive whose guarantees match those requirements. Reject Orleans when the problem is primarily shared-memory computation, a finite batch, relational querying, or global coordination with few independent entities.

Inspect package versions for version-sensitive work. Orleans `10.3.1` ships `Microsoft.Orleans.DurableJobs*` and `Microsoft.Orleans.Journaling*` as `10.3.1-alpha.1`; treat them as experimental until that status changes.

Orleans 10.3 makes Newtonsoft storage enforce the type allow-list, changes RPC telemetry keys, and requires custom grain-context activators to apply configurators before construction. Prefer generated/allowed types over permissive JSON and update affected tests, dashboards, and activators together. `10.3.1` services analyzer contract identities and documents placement hints.

Mental Model

• A **grain** is a virtual actor with stable identity, behavior, and optional state, not a process, row, DTO, controller, or background job.
• A **grain reference** is a location-transparent address. Getting a reference does not create a durable record or prove that an activation exists.
• An **activation** is an ephemeral in-memory execution instance. Orleans creates, places, moves, deactivates, and recreates it. Never equate activation lifetime with entity lifetime.
• A normal grain has at most one activation in the cluster by default and processes turns one at a time. This makes the grain a natural owner of per-identity invariants.
• A **silo** hosts activations. Silos form a cluster; external clients or a co-hosted `IGrainFactory` call grains.
• Calls are asynchronous messages even when they look like C# method calls. Network failure, timeout, serialization, retries, and duplicate side effects still matter.
• A grain can model a **digital twin** when its identity and behavior map to a real or domain entity; that is a use case, not every grain's definition.
• Orleans gives logical ownership and turn-based execution. It does not make external side effects transactional, turn arbitrary data into a queryable database, or provide exactly-once execution by default.

Workflow

1. Inspect the solution, Orleans version, hosting topology, grain interfaces, providers, and tests. 2. Identify domain identities and invariants. Prefer many independent, bounded entities over global coordinator grains. 3. Choose state, communication, and time-based work from the tables below. 4. Keep default non-reentrant scheduling and placement until a measured requirement justifies a change. 5. Configure providers independently and test the failure model the design depends on.

Choose the State Owner

| Primitive | Purpose | Choose it when | Do not use it as | |---|---|---|---| | Activation fields | Fast, temporary state for one activation | The value is derived, cached, disposable, or safe to rebuild after deactivation/failure | Durable truth | | `IPersistentState<T>` | Durable current state owned by one grain identity | The grain needs a bounded snapshot loaded on activation and explicitly written after commands | A general query database or cross-grain table | | External database/repository | Queryable, indexed, relational, bulk, shared, or externally owned data | The system needs joins, search, reporting, set-based updates, independent access, or an existing system of record | A replacement for grain ownership when serialized per-entity decisions are still required | | Grain plus database/read model | Separate command ownership from query/storage concerns | A grain owns invariants and a small control snapshot while a database owns large records, history, projections, or reporting | Two competing sources of truth without an explicit contract | | `JournaledGrain<TState,TEvent>` | Persist domain events and reconstruct state | Audit history, business-event replay, log consistency, or multi-cluster event-sourced replication is a requirement | A default persistence choice for ordinary CRUD state | | `Orleans.Journaling` durable states | Replay durable collection/value operations through a journal | The experimental 10.3 journaling model, durable collections, or durable completion state solves a measured need | Stable default persistence; it is distinct from `JournaledGrain` business event sourcing | | `ITransactionalState<T>` | ACID, serializable all-or-nothing changes across transactional grain state | A short operation must atomically update multiple grain-owned states and compensation is unacceptable | Long-running workflows or atomicity with arbitrary external systems | | Saga/process manager | Durable progress with compensation across steps and external systems | Work is long-running, spans services, waits for events, or cannot share one transaction | Instant atomic commit |

Grain State Versus a Database

Use grain state for bounded current state and per-identity invariants. Use a database/read model for joins, search, reporting, bulk work, history, shared access, or an external system of record. When using both, define one authority per field and recovery rules.

Never query or mutate another grain's persistence record behind the grain, or expose provider stora