Platform of Trust Platform Design Guide
Developer PortalAPI DocsOntology
  • Purpose and scope
  • Developer eXperience Strategy
  • Suggest changes
  • API Requests
    • Request validation
    • Response codes used
    • Error handling
  • General API guidelines
    • Performance
    • Documentation
    • Data models used
    • Use HATEOAS
    • Use HTTP Methods
    • Dates and time
    • Naming conventions
    • Pagination, partial response and sorting
    • API Testing
  • API Headers
    • Mandatory elements & behaviour
    • Rate limiting information
  • API Security
    • Authentication and authorization
    • SSL everywhere - all the time
  • API versioning guidelines
    • Versioning Rules
    • Breaking changes
    • Non-breaking changes example
    • Retirement process
    • Add new API to documentation
    • Add new endpoint
  • API Migration Policies
    • Deprecating API
    • Sunsetting API
    • Blackout Testing
    • Migration Email Template
    • API Blackout Test Email Template
    • API Deprecation Email
    • Deprecating an older API
  • Ontologies
    • About ontologies
    • Web Ontology Language, OWL
    • Ontology editor
    • Edit ontology
    • Add new subclass
    • Naming convention logic
    • Extending the ontology
  • Design Guideline
    • Colors
    • Typography
    • Grids and Space
    • Input forms - Text field
    • Input forms - Text area
    • Buttons
    • Checkbox
    • Radio buttons
    • Date picker
    • Form control - Single select
    • Toggle
    • Pagination
    • Status pills
    • Tables
    • Effects
    • Dialogues
Powered by GitBook
On this page
  • Minimum response time
  • Metrics

Was this helpful?

  1. General API guidelines

Performance

Minimum response time

You should design REST APIs in a way which does not enable inefficient use of it. In other words enabling full datasource replication with single API call with tens of millions of objects is not allowed. To transfer big datamasses you should seek other options or use pagination.

As a rule of thumb,

  • Platform of Trust REST APIs should in 99% of calls respond in under 200 ms preferably even faster.

  • GraphQL API should respond in under x00 ms.

Reasoning for the above limits is that the customer experience of the applications built on top of our platform depend partially on our APIs responsiveness. Our service including APIs must be efficient and reliable.

Metrics

Platform metrics and how it is measured

PreviousError handlingNextDocumentation

Last updated 6 years ago

Was this helpful?