1|# 05 - AWS Cloud Infrastructure — Home
     2|**Purpose:** AWS architecture, networking, and cost optimization for AI workloads.
     3|
     4|## Overview
     5|
     6|This section covers our AWS cloud infrastructure setup for scalable AI services, including:
     7|- VPC networking architecture
     8|- EC2 instance optimization for GPU workloads
     9|- S3 data pipeline automation
    10|- Cost optimization strategies
    11|
    12|## Infrastructure Components
    13|
    14|### **AWS Architecture**
    15|- **US-East-1 (N. Virginia):** Primary production environment
    16|- **Multi-AZ deployment:** High availability for critical services
    17|- **VPC networking:** Isolated subnets for compute, storage, and monitoring
    18|
    19|### **Compute Resources**
    20|- **p4d.24xlarge:** A100-based instances for training
    21|- **g5.xlarge:** A10 instances for inference
    22|- **m5.large:** CPU-based instances for web services
    23|- **Spot instances:** Cost savings for non-critical workloads
    24|
    25|### **Storage**
    26|- **S3 buckets:** Model weights, datasets, backup archives
    27|- **EBS volumes:** High-performance storage for active VMs
    28|- **Glacier:** Cold storage for historical data
    29|
    30|## Cost Optimization
    31|
    32|- **Reserved Instances:** 1-3 year commitments for baseline workloads
    33|- **Spot Instances:** 70% savings for fault-tolerant tasks
    34|- **S3 Lifecycle Policies:** Automatic tiering to Glacier
    35|- **CloudWatch Alarms:** Budget alerts and anomaly detection
    36|
    37|## Related Documents
    38|
    39|- **[Qwen3.5-MoE Setup](../02\ -\ TECHNICAL\ INFRASTRUCTURE/04\ -\ Qwen3.5\ Model\ Setup/)** — Model deployment strategies
    40|
    41|---
    42|
    43|*Our AWS infrastructure supports our SMB AI services and agent deployments with maximum efficiency.*
    44|