Skip to content

Latest commit

 

History

History
38 lines (27 loc) · 2.45 KB

ADR-01-Event Driven.md

File metadata and controls

38 lines (27 loc) · 2.45 KB
Raw

ADR-01: Event-Driven Architecture

Date:

2024-09-29

Status:

Accepted

Context:

The ClearView platform must efficiently process candidate actions like uploading resumes, updating profiles, searching for jobs, and accessing career tips. These tasks must be handled in near real-time to ensure a smooth and efficient user experience. Similarly, the platform must also handle employer actions, such as searching for candidates, updating job postings, and exploring tips, with the same real-time responsiveness to meet business demands.

Decision:

We will implement an Event-Driven Architecture (EDA) to decouple services and facilitate real-time communication across the platform's services and components.

Drivers:

  • Scalability: The platform needs to support growing volumes of candidates, employers, resumes, jobs, and job matching requests.
  • Decoupling: Services like job management, resume processing, and job matching should function independently, enabling easier updates and enhanced fault tolerance.
  • Real-Time Processing: Both candidates and employers expect immediate responses after performing actions on the platform.

Scenarios:

  • When a candidate uploads a resume, the application generates an event (e.g., ResumeUploadedEvent) with the file details.
  • This event is published to an event broker, where other components (e.g., Resume Parser, Job Matcher) subscribe to and process the event accordingly.
  • After the ResumeUploadedEvent is handled, subsequent events, such as ResumeParsed and StoryCreated, can trigger further processing by related services.

Consequences:

Pros:

  • Scalability: Individual services can scale independently, handling varying loads without impacting others.
  • Loose Coupling: Microservices are decoupled, allowing for independent development, updates, and deployments.
  • Improved Responsiveness: User actions are processed in near real-time, improving the overall user experience.
  • Enhanced Fault Tolerance: The system can continue functioning even if some services experience temporary downtime.

Cons:

  • Increased Complexity: The overall system becomes more complex, requiring robust monitoring and logging infrastructure.
  • Eventual Consistency: Data consistency across services may not be immediate, requiring careful management of business processes.
  • Learning Curve: Development teams may need training to fully understand and utilize EDA principles and tools effectively.