Docs
Launch GraphOS Studio
You're viewing documentation for a previous version of this software. Switch to the latest stable version.

Query plans

How your gateway plans operations across subgraphs


⚠️ You definitely don't need to understand the details of query plans to get started with Apollo Federation! This information is provided primarily for advanced debugging purposes.

Whenever your gateway receives an incoming GraphQL , it needs to figure out how to use your s to populate data for each of that operation's s. To do this, the gateway generates a query plan:

Parallel
Fetch (reviews)
Fetch (users)
Fetch (hotels)
Flatten (latestReviews,[],hotel)

A query plan is a blueprint for dividing a single incoming into one or more s that are each resolvable by a single . Some of these operations depend on the results of other s, so the query plan also defines any required ordering for their execution.

Example graph

Let's say our federated graph includes these s:

# Hotels subgraph
type Hotel @key(fields: "id") {
id: ID!
address: String!
}
type Query {
hotels: [Hotel!]!
}
# Reviews subgraph
extend Type Hotel @key(fields: "id") {
id: ID! @external
reviews: [Review!]!
}
type Review {
id: ID!
rating: Int!
description: String!
}

Based on these s, clients can execute the following query against our gateway:

query GetHotels {
hotels { # Resolved by Hotels subgraph
id
address
reviews { # Resolved by Reviews subgraph
rating
}
}
}

This query includes s from both the Hotels and the Reviews subgraph. Therefore, the gateway needs to send at least one query to each subgraph to populate all requested fields.

Take a look at the gateway's query plan for this query:

This syntax probably looks confusing! 🤔 Let's break it down.

Structure of a query plan

A query plan is defined as a hierarchy of nodes that looks like a JSON or GraphQL document when serialized.

At the top level of every query plan is the QueryPlan node:

QueryPlan {
...
}

Each node defined inside the QueryPlan node is one of the following:

NodeDescription
FetchTells the gateway to execute a particular operation on a particular subgraph.
ParallelTells the gateway that the node's immediate children can be executed in parallel.
SequenceTells the gateway that the node's immediate children must be executed serially in the order listed.
FlattenTells the gateway to merge the data returned by this node's child Fetch node with data previously returned in the current Sequence.

Each of these is described in further detail below.

Fetch node

A Fetch node tells the gateway to execute a particular GraphQL on a particular . Every query plan includes at least one Fetch node.

# Executes the query shown on the "books" subgraph
Fetch(service: "books") {
{
books {
title
author
}
}
},

The node's body is the to execute, and its service indicates which to execute the against.

In our example graph above, the following query requires data only from the Hotels :

query GetHotels {
hotels {
id
address
}
}

Because this doesn't require orchestrating operations across multiple s, the entire query plan contains just a single Fetch node:

QueryPlan {
Fetch(service: "hotels") {
{
hotels {
id
address
}
}
},
}

The Fetch node uses a special syntax when it's resolving a reference to an entity across s. For details, see Resolving references with Flatten.

Parallel node

A Parallel node tells the gateway that all of the node's immediate children can be executed in parallel. This node appears in query plans whenever the gateway can execute completely independent s on different s.

Parallel {
Fetch(...) {
...
},
Fetch(...) {
...
},
...
}

For example, let's say our federated graph has a Books and a Movies subgraph. And let's say a client executes the following query to fetch separate lists of books and movies:

query GetBooksAndMovies {
books {
id
title
}
movies {
id
title
}
}

In this case, the data returned by each does not depend on the data returned by any other . Therefore, the gateway can query both subgraphs in parallel.

The query plan for the looks like this:

Sequence node

A Sequence node tells the gateway that the node's immediate children must be executed serially in the order listed.

Sequence {
Fetch(...) {
...
},
Flatten(...) {
Fetch(...) {
...
}
},
...
}

This node appears in query plans whenever one 's response depends on data that first must be returned by another . This occurs most commonly when a query requests s of an entity that are defined across multiple s.

As an example, we can return to the GetHotels query from our example graph:

query GetHotels {
hotels { # Resolved by Hotels subgraph
id
address
reviews { # Resolved by Reviews subgraph
rating
}
}
}

In our example graph, the Hotel type is an entity. Hotel.id and Hotel.address are resolved by the Hotels , but Hotel.reviews is resolved by the Reviews . And our Hotels subgraph needs to resolve first, because otherwise the Reviews doesn't know which hotels to return reviews for.

The query plan for the looks like this:

As shown, this query plan defines a Sequence that executes a Fetch on the Hotels before executing one on the Reviews . (We'll cover the Flatten node and the second Fetch's special syntax next.)

Flatten node

A Flatten node always appears inside a Sequence node, and it always contains a Fetch node. It tells the gateway to merge the data returned by its Fetch node with data that was previously Fetched during the current Sequence:

Flatten(path: "hotels.@") {
Fetch(service: "reviews") {
...
}
}

The Flatten node's path tells the gateway at what position to merge the newly returned data with the existing data. An @ element in a path indicates that the immediately preceding path element returns a list.

In the snippet above, the data returned by the Flatten's Fetch is added to the Sequence's existing data within the objects contained in the hotels list .

Expanded example

Once again, let's return to the GetHotels query on our example graph:

query GetHotels {
hotels { # Resolved by Hotels subgraph
id
address
reviews { # Resolved by Reviews subgraph
rating
}
}
}

The query plan for this first instructs the gateway to execute this query on the Hotels :

{
hotels {
id
address
__typename # The gateway requests this to resolve references (see below)
}
}

At this point, we still need review-related information for each hotel. The query plan next instructs the gateway to query the Reviews for a list of Hotel objects that each have this structure:

{
reviews {
rating
}
}

Now, the gateway needs to know how to merge these Hotel objects with the data it already fetched from the Hotels . The Flatten node's path tells it exactly that:

Flatten(path: "hotels.@") {
...
}

In other words, "Take the Hotel objects returned by the Reviews and merge them with the Hotel objects in the top-level hotels returned by the first query."

When the gateway completes this merge, the resulting data exactly matches the structure of the client's original query:

{
hotels {
id
address
reviews {
rating
}
}
}

Resolving references with Flatten

Like Sequence nodes, Flatten nodes appear whenever one 's response depends on data that first must be returned by another . This almost always involves resolving entity s that are defined across multiple subgraphs.

In these situations, the Flatten node's Fetch needs to resolve a reference to an entity before fetching that entity's s. When this is the case, the Fetch node uses a special syntax:

Flatten(path: "hotels.@") {
Fetch(service: "reviews") {
{
... on Hotel {
__typename
id
}
} =>
{
... on Hotel {
reviews {
rating
}
}
}
},
}

Instead of containing a GraphQL , this Fetch node contains two GraphQL fragments, separated by =>.

  • The first is a representation of the entity being resolved (in this case, Hotel). Learn more about entity representations.
  • The second contains the entity s and subfields that the gateway needs the to resolve (in this case, Hotel.reviews and Review.rating).

When the gateway sees this special Fetch syntax, it knows to query a 's Query._entities field. This is what enables a to provide direct access to any available fields of an entity.

Now that you've learned about each query plan node, take another look at the example query plan in Example graph to see how these nodes work together in a complete query plan.

Viewing query plans

You can view the query plan for a particular in any of the following ways:

Outputting query plans from the gateway

Your gateway can output the query plan for each incoming as it's calculated. To do so, add the following to the file where you initalize your ApolloGateway instance:

  1. Import the serializeQueryPlan function from the @apollo/query-planner library:

    const {serializeQueryPlan} = require('@apollo/query-planner');
  2. Add the experimental_didResolveQueryPlan option to the object you pass to your ApolloGateway constructor:

    const gateway = new ApolloGateway({
    experimental_didResolveQueryPlan: function(options) {
    if (options.requestContext.operationName !== 'IntrospectionQuery') {
    console.log(serializeQueryPlan(options.queryPlan));
    }
    }
    });

    The value you provide for this option is a function that's called every time the gateway generates a query plan. The example function above logs the generated query plan for every except for queries (such as those sent periodically by tools like the Apollo Studio ). You can define any logic you want to log query plans or otherwise interact with them.

    For all available options passed to your function, see the source.

Previous
Monitoring
Next
Error codes
Edit on GitHubEditForumsDiscord