API / Apollo Gateway Interview questions
Explain the execution flow of a federated query spanning three subgraphs?
Take a query asking for a product's name, its reviews, and each reviewer's display name — with Products, Reviews, and Users each owning a different piece.
sequenceDiagram
participant C as Client
participant G as Gateway
participant P as Products subgraph
participant R as Reviews subgraph
participant U as Users subgraph
C->>G: query { product(id) { name reviews { body author { name } } } }
G->>P: fetch product name + id
P-->>G: { id, name }
G->>R: _entities([{__typename:"Product", id}]) for reviews
R-->>G: reviews list with body + authorId per review
G->>U: _entities([{__typename:"User", id: authorId}, ...]) for names
U-->>G: user names for each authorId
G->>G: stitch results into one response tree
G-->>C: final merged JSON response
Note that the call to Users happens as a single batched _entities request covering every reviewer needed for that query, not one call per review. Only after all three subgraph calls resolve does the gateway assemble the final tree, matching exactly the shape the client originally asked for.
In this example, which subgraph is queried first, and why?
How many calls does the gateway make to the Users subgraph to get all reviewer names?
More Related questions...
