Faasera Deployment Guide
This guide explains how to deploy Faasera components across various environments, including serverless cloud platforms,
SDK-based pipelines, and the full visual UI stack. It provides deployment patterns, prerequisites, and operational best
practices.
Deployment Modes Overview
| Mode |
Description |
Ideal For |
| Cloud Functions |
Lightweight masking, profiling, and validation via serverless functions |
Event-driven processing, microservices |
| SDK Integration |
Java / PySpark SDK for Spark and Databricks environments |
Batch/streaming data pipelines |
| REST API |
Host the Faasera engine with REST endpoints |
CI/CD automation, centralized control |
| Full UI Platform |
Visual orchestration and management platform |
End-to-end compliance operations |
| Plugin Mode |
Use with ETL tools like NiFi, ADF, Airflow |
Drag-and-drop pipeline integration |
Cloud Function Deployment
- AWS Lambda
- Azure Functions
- GCP Cloud Functions
Requirements:
- Runtime: Java 17+, Python 3.8+
- Memory: 512MB+ (depends on data volume)
- Trigger: HTTP, EventBridge, Blob triggers, S3, etc.
Example: AWS Lambda
# Deploy via AWS CLI
aws lambda create-function --function-name FaaseraFunction --runtime java11 --handler ai.faasera.lambda.
FaaseraHandler --zip-file fileb://build/libs/faasera.<build>.jar --role
arn:aws:iam::<account>:role/FaaseraLambdaRole
SDK Deployment (Java / PySpark)
Use Cases:
- Databricks Notebooks
- Spark Streaming Pipelines
- Airflow DAG with custom UDFs
Java SDK
- Add dependency to
pom.xml or build.gradle
- Initialize Faasera engine via builder
- Pass policy, dataset, and config object
PySpark SDK
pip install faasera-sdk- Import
FaaseraMaskingEngine
- Pass Spark DataFrame and masking policy
REST API Deployment
Quickstart:
- Package: Spring Boot or Micronaut-based application
- Endpoints:
/profile, /mask, /validate, /generate, /risk
- Run:
java -jar faasera-api-server.jar
Recommended Setup:
- Reverse proxy via NGINX
- OAuth2 or token-based auth
- Auto-scaling via Kubernetes (optional)
Architecture:
- Backend: Java (Spring Boot)
- Frontend: React + REST/GraphQL
- Database: PostgreSQL
- Orchestration: Docker Compose or Kubernetes
Quickstart (Docker Compose):
Recommended:
- Host behind HTTPS ingress
- Configure API gateway for rate limiting
- Use managed PostgreSQL (e.g., RDS)
Plugin Deployment
Available Plugins:
- Apache NiFi
- Azure Data Factory (ADF)
- Apache Airflow
- Control-M
Usage:
- Drag Faasera transform step into flow
- Configure policy file, connection, masking mode
- Monitor via native ETL dashboard
Licensing & Security
- Functions and SDKs support offline license tokens
- Optional rate-limiting, API keys, and RBAC controls
- Encryption in-transit (TLS) and at-rest (AES/FPE)
- Logs and usage metrics can be pushed to SIEM/Splunk
Best Practices
| Area |
Recommendation |
| Scalability |
Use serverless for burst loads, SDKs for large volumes |
| Monitoring |
Enable Prometheus/Grafana or use logs with ELK/Splunk |
| Compliance |
Store policies in version-controlled repo (e.g., Git) |
| Auditing |
Retain validation logs and masking reports for 12+ months |
| Security |
Rotate encryption keys and API tokens regularly |