DataTerminology.comNEW
Cloud model where providers manage servers and autoscale per request, letting teams focus on code and events.
DESCRIPTION
Serverless computing abstracts servers, capacity, and patching from developers. Functions and managed services scale automatically based on demand and charge per use. It accelerates delivery of event-driven applications, data pipelines, and APIs. Operational concerns shift to cold starts, observability, limits, and vendor lock-in. Patterns include FaaS, event routing, and step orchestration. Many architectures combine serverless with containers for flexibility and portability.
ALTERNATIVE DEFINITIONS
Wikipedia
Serverless computing is a cloud-computing execution model in which the cloud provider runs the server, and dynamically manages the allocation of machine resources.
Stresses provider-managed infrastructure and dynamic resource allocation.
Microsoft
Serverless computing enables developers to build applications faster by eliminating the need to manage infrastructure.
Emphasizes developer productivity from eliminating infrastructure management.
Google Cloud
Serverless computing lets you build and run applications without having to manage servers.
Highlights the no-servers-to-manage promise succinctly.
RELATED TERMS
Function as a Service Backend as a Service Event-Driven Architecture Microservices Containers Initialization Managed Services Stateless Functions
COMPETING TERMS
Containers on Kubernetes Virtual Machines
KNOWLEDGE AREAS
Cloud Infrastructure Platform & Tooling Data Engineering AI
TARGET AUDIENCE
Software Developer Data Engineer AI Engineer Cloud Architect MLOps Engineer
EXTERNAL SOURCES
- Wikipedia - Serverless computing
- AWS - Overview of AWS serverless services
- Google Cloud - Google Cloud serverless overview
- Microsoft - Azure serverless architecture guidance
TOP QUESTIONS
When is Serverless Computing a good fit?
It excels for event-driven workloads, unpredictable traffic, and bursty tasks where per-request scaling and pricing shine.
Category: Guidance
What are the biggest risks with Serverless Computing?
Cold starts, vendor lock-in, and service limits can impact latency and design. Observability can be challenging across managed boundaries.
Category: Risk
How does Serverless Computing affect cost?
You pay per execution or consumption, often reducing idle costs. High-throughput steady workloads may be cheaper on containers or VMs.
Category: Cost
Can I run ML inference on Serverless Computing?
Yes, for moderate payloads and latency targets. For GPU-bound or large models, consider container services with accelerators.
Category: Application
How do I mitigate cold starts in Serverless Computing?
Use provisioned concurrency, smaller packages, and keep-alive traffic. Choose runtimes with fast startup.
Category: Best Practices
Is Serverless Computing always stateless?
Functions are stateless between invocations; persist state externally in databases, caches, or queues.
Category: Concepts
How does observability work in Serverless Computing?
Leverage managed logs, traces, and metrics; propagate context through events and use centralized tracing tools.
Category: Observability
What are common serverless architectures for data pipelines?
Event ingestion, stream processing, function-based ETL, and scheduled orchestration integrate with managed storage and analytics.
Category: Patterns
How portable are Serverless Computing workloads?
They can be portable with open APIs and frameworks, but proprietary triggers and services increase lock-in risk.
Category: Opinion