Global Accelerator

0

AWS Global Accelerator — From Basics to Advanced

For applications requiring low latency, high availability, and global reach, AWS Global Accelerator is the ideal solution. It provides static IP addresses that act as fixed entry points to your application and intelligently routes traffic over the AWS global network to improve performance.

This blog covers basics, components, creation, traffic routing, health checks, and advanced configurations.


1️⃣ What is AWS Global Accelerator?

Definition:
AWS Global Accelerator is a network service that improves availability and performance of your applications by directing user traffic to optimal AWS endpoints like ALB, NLB, or EC2 instances using the AWS global network instead of the public internet.

Key Benefits:

  • Static IP addresses for your application (simplifies DNS)

  • Automatic traffic routing to the closest healthy endpoint

  • Reduces latency using AWS global backbone network

  • Supports failover and high availability

  • Works with ALB, NLB, and EC2 endpoints

Example Use Case:

  • A web application hosted in us-east-1 and eu-west-1: Global Accelerator automatically directs US users to US endpoint and EU users to EU endpoint with the lowest latency.


2️⃣ Core Components of Global Accelerator

Component Role
Global Accelerator The main service; provides static IP addresses.
Accelerator IP Addresses Two static IPv4 addresses that serve as entry points.
Listeners Listen for traffic on specific ports/protocols (TCP/UDP).
Endpoint Groups Groups of endpoints in a region; control traffic distribution.
Endpoints Resources that serve traffic (ALB, NLB, EC2).
Health Checks Monitors endpoint health and routes traffic to healthy endpoints.

3️⃣ How Traffic Flows

  1. Users connect to the static Global Accelerator IP addresses

  2. Traffic enters the closest AWS edge location

  3. AWS routes the traffic over the AWS backbone network to the optimal regional endpoint

  4. Traffic reaches your application hosted on ALB, NLB, or EC2

Benefit: Users get low-latency, reliable access even during regional failures.


4️⃣ Step-by-Step Global Accelerator Setup

Step 1: Create a Global Accelerator

  1. Go to AWS Global Accelerator → Create Accelerator

  2. Configure:

    • Name: MyGlobalAccelerator

    • Type: Standard (for TCP/UDP applications)

    • Optional: Custom IP addresses

You will get two static IPv4 addresses automatically.


Step 2: Create a Listener

  • Protocol: TCP or UDP (choose as per your application)

  • Port: Specify port(s) (e.g., 80, 443)


Step 3: Configure Endpoint Groups

  • Assign region(s) where your application is deployed

  • Optional: Configure traffic dial to control % of traffic per region

  • Enable Health Checks:

    • Protocol: TCP/HTTP/HTTPS

    • Port: Target port

    • Path (for HTTP/HTTPS): /health


Step 4: Add Endpoints

  • Add ALB, NLB, or EC2 instances in the selected region

  • For multiple endpoints in a region, you can set weighting to distribute traffic


Step 5: Save & Test

  • Global Accelerator is now active

  • Users connect to static IP addresses

  • Test failover by stopping an endpoint — traffic automatically routes to healthy endpoints


5️⃣ Advanced Features

  1. Traffic Dial: Control % of traffic going to each region

  2. Endpoint Weighting: Split traffic among multiple endpoints within a region

  3. Cross-Region Failover: Automatically redirects traffic to healthy regions

  4. Custom Domain Names: Point your DNS (Route 53 or external) to the static IP addresses

  5. Integrated with ALB/NLB: Leverages Layer 4/Layer 7 load balancers for backend distribution


6️⃣ Monitoring & Metrics

  • CloudWatch Metrics:

    • HealthyEndpointCount — Number of healthy endpoints

    • ClientIPv4Count — Unique clients served

    • ProcessedBytes — Bytes sent/received

  • Access Logs: Can enable Flow Logs for endpoint traffic analysis


7️⃣ Best Practices

  1. Use Global Accelerator for globally distributed applications

  2. Always enable health checks for failover

  3. Combine with ALB/NLB to leverage both Layer 7 & Layer 4 routing

  4. Use Route 53 or external DNS pointing to static IPs for seamless domain integration

  5. Configure traffic dial for controlled rollouts or testing


8️⃣ Comparison Table: ALB vs NLB vs Global Accelerator

Feature ALB NLB Global Accelerator
OSI Layer 7 4 3/4
Protocol HTTP/HTTPS TCP/UDP/TLS TCP/UDP
Routing Host/Path/Header/Query Source IP / Port Regional failover & proximity-based
Static IP No Yes (Elastic IP optional) Yes (default two IPs)
TLS Termination Yes Yes No (uses backend TLS)
Session Stickiness Cookies Source IP No
Global Failover No No Yes
Use Case Web apps, microservices Real-time apps, high-performance Globally distributed applications

9️⃣ Conclusion

AWS Global Accelerator improves application availability and performance globally:

  • Provides static IP addresses as fixed entry points

  • Routes traffic over the AWS backbone network to reduce latency

  • Automatically directs traffic to healthy endpoints in any region

  • Integrates seamlessly with ALB, NLB, and EC2 for flexible application architecture

Use Global Accelerator when you want global performance, automatic failover, and static entry points for your applications.

Leave a Reply

Your email address will not be published. Required fields are marked *

Scroll to Top