Data & Beyond REST — Chapter Quiz

Question 1

1

Why use a connection pool (e.g., pg.Pool)?

ASmaller bundle sizeBReuse connections — opening one is expensive (~10ms each)CType safetyDCSRF protection

Pools keep a fixed number of connections open and reuse them across requests.

Question 2

2

What's the right way to handle a 500 MB file upload?

ABuffer the whole thing in memoryBUse a presigned S3 URL — client uploads directly, your server never touches the bytesCSave it to your DBDReject the request

Presigned URLs scale infinitely; your Node process isn't a bottleneck.

Question 3

3

Server-Sent Events are bi-directional.

TTrueFFalse

SSE is server → client only. WebSockets are bi-directional.

Question 4

4

Why verify webhook signatures BEFORE parsing the body as JSON?

ASpeedBYou need the exact raw bytes to compute HMAC; express.json() discards themCIt is required by ExpressDNo specific reason

Use express.raw({ type: "application/json" }) on webhook routes so you have the bytes for signature verification.

Question 5

5

What does BullMQ use as its underlying store?

APostgreSQLBRedisCMongoDBDIn-memory only

BullMQ stores queues in Redis. Workers poll Redis for jobs.

Question 6

6

The fastest way to handle a slow webhook is to verify, enqueue, and reply ___.

Verify → enqueue → reply 200. Process the actual work in a background worker so the provider doesn't time out.

Question 7

7

When does GraphQL beat REST?

AAlwaysBWhen one UI needs deeply nested data and you have many client typesCWhen you want easy HTTP cachingDPublic APIs

REST is the safe default. GraphQL shines when client variety and field-level granularity matter more than HTTP caching.

Question 8

8

What's the DataLoader pattern for?

ALoading TypeScript typesBBatching and caching DB lookups in a request to avoid N+1 queriesCValidating inputDCaching responses

DataLoader collects per-request findById calls and turns them into one batched IN(...) query.

Test what you learned.