Backend Development 12 min read

Design and Implementation of JD Daojia Open Platform Services

The article presents a comprehensive overview of JD Daojia's Open Platform, detailing its service lifecycle—from platform introduction, service design, publishing, and authorization to state management and traffic‑control mechanisms—while explaining the underlying RPC‑based architecture and fixed‑window rate‑limiting algorithm.

Dada Group Technology
Dada Group Technology
Dada Group Technology
Design and Implementation of JD Daojia Open Platform Services

Platform Introduction : JD Daojia Open Platform is an open service integration platform for merchants and third‑party developers, offering end‑to‑end business‑scenario solutions, automatic adaptation of integration schemes, and one‑hour delivery services for a wide range of goods.

Service Design : The core of the platform is its service cluster. Services follow a lifecycle of definition → implementation → iteration → deprecation. Design goals emphasize stability, usability (clear request parameters and error codes), security (monitoring and authorization), and extensibility (smooth upgrades and seamless switching). Documentation is generated automatically and delivered via a portal system to reduce communication overhead.

Service Publishing : All external services are created in a service‑publishing system, which produces message‑protocol specifications used by the API gateway and portal documentation. Publishing a new service updates the portal instantly, and iterative updates are handled by adjusting the protocol without downtime.

Service Authorization : Services are grouped by integration schemes. Once a scheme is authorized for a merchant, the merchant gains access to all services within that scheme. New services require explicit inclusion in the relevant scheme to become accessible.

State Management : Each service has a lifecycle state—development, testing, released, hidden, or disabled. The hidden state indicates that a service no longer accepts new merchants but remains compatible for existing ones.

Traffic Control : The platform handles millions of calls per minute and employs global and local rate‑limiting using a fixed‑window counter algorithm backed by Redis. Global limits protect downstream services, while per‑machine limits safeguard the platform itself. The fixed‑window method was chosen for its simplicity and low memory footprint despite occasional burst spikes.

Summary : The article outlines the overall design of JD Daojia's Open Platform, covering service design standards, publishing workflow, authorization model, state management, and traffic‑control algorithm selection, aiming to ensure service stability and provide practical insights for readers.

microservicesbackend developmentAPI Gatewaytraffic controlOpen Platformservice architecture
Dada Group Technology
Written by

Dada Group Technology

Sharing insights and experiences from Dada Group's R&D department on product refinement and technology advancement, connecting with fellow geeks to exchange ideas and grow together.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.