Serverless Computing
Concepts and terms for serverless architectures.
The core ideas of serverless computing: functions as a service, event-driven execution, cold starts, statelessness, API gateways and pay-per-use billing, plus the operational concepts that matter in practice: concurrency, timeouts, idempotency, scale to zero and lock-in.
Use it when
- You are deciding whether a workload fits a serverless model.
- You need the vocabulary for reading a serverless platform's documentation.
Watch out for
- Serverless suits bursty, event-driven work; steady high-throughput workloads can be cheaper on long-running servers.
- Limits on duration, memory and concurrency are provider-specific and change over time.
The data
13 rows
| Concept | Description |
|---|---|
| Function as a Service | Event-driven functions executed in ephemeral containers. |
| Event-Driven | Functions triggered by events such as HTTP requests or queue messages. |
| Cold Start | Latency from initializing a function runtime after idle time. |
| Stateless | Functions do not retain state between invocations. |
| API Gateway | Managed service routing HTTP requests to functions. |
| Pay-Per-Use | Billing based on actual execution time and request count. |
| Provisioned Concurrency | Keeping a set number of function instances warm to avoid cold-start latency, at a fixed cost. |
| Function Timeout | The maximum time one invocation may run before the platform stops it. |
| Idempotency | Designing handlers so repeated delivery of the same event has the same effect as one, since many triggers are at-least-once. |
| Scale to Zero | Running no instances, and paying nothing, when there is no traffic. |
| Edge Functions | Functions executed in many locations close to users, typically with tighter runtime limits. |
| Vendor Lock-In | Reliance on proprietary triggers, SDKs and managed services that makes moving to another provider costly. |
| Container-Based Functions | Serverless platforms that run a container image on demand and scale it automatically, for example Google Cloud Run. |
Fields
The shape of this dataset, read from the data itself: every column, its type and a real value. Agents can use this to plan a query without fetching the whole document.
| Section | Field | Type | Example |
|---|---|---|---|
| data | concept | string | Function as a Service |
| data | description | string | Event-driven functions executed in ephemeral co… |
Fetch the same bytes
The static files contain the same reference rows as the API, with no API rate limit. Live metadata, such as view counts, may differ. Use the API for content negotiation; use the static files for a cacheable snapshot.
curl "https://yjtoon.com/api/dataset/serverless-computing?format=toon"
const res = await fetch( "https://yjtoon.com/static-data/dataset/serverless-computing.toon" ); const toon = await res.text();
Rate limit: 120 requests per minute per IP, no key and no signup. API reference →