Large-scale Traffic
Handling Live Streaming
Infrastructure Development

AWS

OliveYoung Live Commerce

Project Logo
- Access URL: https://www.olcl.shop
- Test ID: user1 / CloudWave!




Project Introduction

  • Multi-channel live service providing 3 or more live channels
  • Stable live service through cloud-based traffic handling
  • Personalized service utilizing AWS SaaS based ML services



1. Technology Stack

  1. Infrastructure

    • Nginx
    • Kubernetes
    • EC2

  2. Development

    • Streamlit
    • SpringBoot
    • Lambda

  3. Database

    • RDS
    • DynamoDB
    • S3

  4. CI/CD

    • GitLab
    • Jenkins
    • ArgoCD

  5. Monitoring

    • Loki
    • Prometheus
    • Grafana
    • CloudWatch

  6. Collaboration

    • Notion
    • Slack




2. Project Duration and Schedule Management

Project Duration

  • August 12, 2024 ~ August 31, 2024

Schedule Management

  • Sharing progress through Notion and Slack
  • Daily morning meetings and Agile task distribution




3. Project Differentiators

  • Central management environment for separation and isolation of development and operational environments
  • Warm Standby DR (RTO: 5min / RPO: 15min)
  • No Public Bastion → Using EC2 Instance Connect Endpoint
  • Closed CI/CD implementation
  • Shared resources between environments
  • Cost reduction to 1/25 level using thumbnails for Rekognize




4. Overall Architecture

Main Architecture




5. IVS - Service Flow

IVS Service Flow




6. Rekognition + Personalize - Service Flow

Rekognition and Personalize Service Flow




7. Monitoring

Grafana Dashboard
- Prometheus + Loki & CloudWatch




8. Alarm

Alarm System




9-1. CI/CD Pipeline

CI/CD Pipeline




9-2. Lambda Pipeline

Lambda Pipeline




10. Load Test

Load Test Results
- 100,000 requests/s Test using ApacheBench & JMeter




11. Improvement Goals

  1. Monitoring Optimization

    • Transition from single Prometheus-based centralized monitoring → Metric management using Thanos (OSS)

  2. Performance Enhancement through Cache Usage

    • RDS - Elastic Cache connection → Improving database access performance

  3. Advanced Log Analysis

    • Using ELK stack (Elasticsearch - Logstash - Kibana) → Strengthening problem response capabilities