Technical Architecture

1 Architecture

Architecture

This architecture is centered on the Softa Application Development Platform, following a top-down layered model.

The top layer is Business Apps, such as CoreHR, Payroll, Talent, Workforce, and Expenses, which directly serve business scenarios and end users.

The middle layer provides platform capabilities and consists of:

Platform Services – common services including IAM, Workflow, Messaging, Metadata, AI, Integration, Audit, File, Data Quality, and Cron, offering unified capabilities for all applications;

Metadata Framework – Base, ORM, and Web components that support model-driven development and rapid application building.

The bottom layer is Infra, including RDBMS, OSS, Index, MQ, and Cache, providing storage, messaging, and caching support.

Together they form a structure of Business Apps → Platform Services → Metadata Framework → Infra, achieving decoupling between business and technology and enabling fast delivery and continuous scalability.

2 Data Architecture

Data Architecture

This architecture follows a model of Governance-Driven + Domain Access + Multi-Model Storage. Metadata acts as the central hub to decouple data sources, domain services, and physical storage.

The structure consists of:

Vertical dimensions: Data Sources → Data Access → Data Domains → Data Storage
Horizontal layers: Metadata & Governance → Domain APIs → Data Domains → Storage

Core principle: All data must be accessed through Domain APIs, governed by metadata, and abstracted from storage.

2.1 Core Layers

(1) Metadata & Governance

Central control layer responsible for:

Data Schema – unified models
Lineage – data flow tracking
Ownership – responsibility definition
Audit Metadata – compliance records

This layer defines how data is understood and managed, not where it is stored.

(2) Domain APIs

The only entry for data access:

Unified domain interface
Shielding storage diversity
Carrying business rules
Collaborating with governance

It is the bridge between governance and applications.

(3) Data Domains

Data is organized into:

Business Data
Master & Reference Data
System Data

Each domain has its own lifecycle and policy.

(4) Data Storage

Multi-model persistence:

RDBMS – structured data
Index – search acceleration
OSS – unstructured objects

Fully abstracted by Domain APIs.

2.2 Consumers and Ingress

Client App
Integrations
Domain Services

Access path: Data Sources → Domain APIs → Governance → Storage

2.3 Characteristics

Governance before access
Single domain entry
Pluggable storage
Traceable & auditable
Clear responsibilities