System Design PrimerLearn How to Design Large-Scale Systems
  • Introduction
  • Performance vs scalability
  • Latency vs throughput
  • Availability vs consistency
  • Consistency patterns
  • Availability patterns
  • Domain name system
  • Content delivery network
  • Load balancer
  • Reverse proxy
  • Application layer
    • RDBMS
    • NoSQL
    • SQL or NoSQL
  • Cache
  • Asynchronism
    • Overview
    • HTTP
    • TCP
    • UDP
    • RPC and REST
  • Security
    • Overview
    • Powers of two
    • Latency numbers
    • Interview questions
    • Real world architectures
    • Company architectures
    • Company engineering blogs
  • Introduction
  • Performance vs scalability
  • Latency vs throughput
  • Availability vs consistency
  • Consistency patterns
  • Availability patterns
  • Domain name system
  • Content delivery network
  • Load balancer
  • Reverse proxy
  • Application layer
    • RDBMS
    • NoSQL
    • SQL or NoSQL
  • Cache
  • Asynchronism
    • Overview
    • HTTP
    • TCP
    • UDP
    • RPC and REST
  • Security
    • Overview
    • Powers of two
    • Latency numbers
    • Interview questions
    • Real world architectures
    • Company architectures
    • Company engineering blogs

Latency vs throughput#

Latency is the time to perform some action or to produce some result.

Throughput is the number of such actions or results per unit of time.

Generally, you should aim for maximal throughput with acceptable latency.

Source(s) and further reading#

  • Understanding latency vs throughput
Performance vs scalability
Availability vs consistency

MIT 2022 © Jeffrey Liu.
Edit this page on GitHub
  • Latency vs throughput
  • Source(s) and further reading