Batching for N+1 problem

A Mesh Gateway that is not properly configured will face the same famous issue as any other GraphQL servers: the N+1 Query problem.

Fortunately, GraphQL Mesh provides a way to batch requests with specific parameters.

Enable requests batching

Our multiple-sources example Gateway:

Will emit multiple requests to the “Authors” API when resolving the nested Book.author field:

query bestSellersByStore {
  stores {
    id
    name
    bookSells {
      sellsCount
      book {
        id
        title
        author {
          id
          name
        }
      }
    }
  }
}

Fortunately, Mesh allows an extra directive-based configuration to provide a “batching” query that will help resolve many record of the same type.

Our current resolver configuration for Book.author is the following:

.meshrc.yaml

sources:
  # …
transforms:
  # …
additionalTypeDefs: |
  # …
  extend type Book {
    author: authors_v1_Author @resolveTo(
      sourceName: "Authors", # Which source does the target field belong to?
      sourceTypeName: "Query", # Which root type does the target field belong to?
      sourceFieldName: "authors_v1_AuthorsService_GetAuthor", # What is the source field name?
      requiredSelectionSet: "{ authorId }",
      # What args does this need to take?
      sourceArgs: {
        "input.id": "{root.authorId}"
      }
    )
  }

Assuming that the “Authors” is exposing a authors_v1_AuthorsService_GetAuthors(input: authors_v1_GetAuthorsRequest_Input) with authors_v1_GetAuthorsRequest_Input being:

input authors_v1_GetAuthorsRequest_Input {
  ids: [String!]!
}

We could update our .meshrc.yaml configuration as follows:

sources:
  # …
transforms:
  # …

# Create a resolver with batching to solve N+1 problem
additionalTypeDefs: |
  # …
  extend type Book {
    author: authors_v1_Author @resolveTo(
      sourceName: "Authors",
      sourceFieldName: "authors_v1_AuthorsService_GetAuthors",
      keyField: "authorId",
      keysArg: "input.ids"
    )
  }

requiredSelectionSet and sourceArgs got replaced by keyField and keysArg:

• keysArg provides the name of the batching primary key argument (input.ids from authors_v1_GetAuthorsRequest_Input)
• keyField indicates which Book selection-set field should be used to provide the ids value

By default, batched results are mapped back to keys by array position: the upstream must return one result per key in key order. Collection-style sources that return rows in their own order (or with gaps / multiples) should also set valueKeyField to the field on each result used for correlation — see Schema Extensions → valueKeyField.

Now, our Mesh Gateway will try to batch calls to the “Authors” API when resolving Book.author:

Request Batching on the gateway level

Mesh also provides a way to batch requests on the gateway level. This is useful when you want to send multiple requests to the gateway in a single HTTP request. This follows Batching RFC

serve:
  batchingLimit: 10 # You have to define an explicit limit for batching