> ## Documentation Index
> Fetch the complete documentation index at: https://docs.apinizer.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Deployment Topologies

> You can explore three basic installation topologies that the Apinizer platform can use according to different requirements: Test/PoC, Professional, and High Availability. You can compare the advantages, usage areas, and requirements of each topology and select the installation model most suitable for you.

The Apinizer platform offers three basic installation topologies suitable for different requirements and scenarios. Each topology has its own advantages, usage areas, and requirements.

## Topology Comparison

<CardGroup cols={2}>
  <Card title="Topology 1: Test and PoC" icon="flask" href="/en/concepts/deployment/topology-test-poc">
    **Test and Proof of Concept**<br />
    Quick installation, low resource requirements
  </Card>

  <Card title="Topology 2: Professional Installation" icon="briefcase" href="/en/concepts/deployment/topology-professional">
    **Production Environments**<br />
    Medium-scale enterprise applications
  </Card>

  <Card title="Topology 3: High Availability" icon="shield-halved" href="/en/concepts/deployment/topology-ha">
    **HA Production**<br />
    Critical business applications, %99.9+ uptime
  </Card>
</CardGroup>

## Quick Comparison Table

| Feature                      | Topology 1<br />(Test/PoC) | Topology 2<br />(Professional) | Topology 3<br />(HA)      |
| ---------------------------- | -------------------------- | ------------------------------ | ------------------------- |
| **Total Server Count**       | 2                          | 7                              | 12                        |
| **Kubernetes Control-Plane** | 1                          | 1                              | 3 (HA)                    |
| **Kubernetes Worker**        | 1                          | 4                              | 3                         |
| **MongoDB**                  | Single Instance            | Single Instance                | Replica Set (3 nodes)     |
| **Elasticsearch**            | Single Node                | Single Node                    | Cluster (3 nodes)         |
| **High Availability**        | x                          | x                              | ✓                         |
| **Usage Purpose**            | PoC, Test                  | Production                     | Production (HA)           |
| **Traffic Capacity**         | \~5M requests/day (1×W4)   | \~20M requests/day (4×W4)      | \~30M requests/day (3×W8) |

## Topology Selection Guide

<AccordionGroup>
  <Accordion title="When Should Topology 1 (Test/PoC) Be Used?">
    **Usage Scenarios:**

    * Proof of Concept (POC) projects
    * Development and test environments
    * Low-traffic applications (\~5M requests/day, 1×W4)
    * Quick installation requirements
    * Limited resources and budget
    * For training and learning purposes

    **Advantages:**

    * Lowest resource requirements
    * Quick installation
    * Low cost
    * Simple management

    **Disadvantages:**

    * Single point of failure risk
    * No high availability
    * Not suitable for production
    * Limited scalability
  </Accordion>

  <Accordion title="When Should Topology 2 (Professional) Be Used?">
    **Usage Scenarios:**

    * Medium-scale production environments
    * Medium-to-high-traffic applications (\~20M requests/day, 4×W4)
    * Basic high availability requirements
    * Situations requiring budget optimization
    * Enterprise applications

    **Advantages:**

    * Load balancing between worker nodes
    * Medium-level resource requirements
    * Suitable for production
    * Flexible scaling

    **Disadvantages:**

    * Limited high availability
    * Database single point of failure risk
    * No geographic distribution
  </Accordion>

  <Accordion title="When Should Topology 3 (High Availability) Be Used?">
    **Usage Scenarios:**

    * Critical production environments
    * High-traffic applications (\~30M requests/day and above, 3×W8)
    * High availability requirements (%99.9+ uptime)
    * Critical business processes
    * Data security and replication requirements

    **Advantages:**

    * High availability (%99.9+)
    * Automatic failover
    * Data replication
    * Load balancing
    * Zero-downtime updates

    **Disadvantages:**

    * High resource requirements
    * Complex installation
    * High cost
    * Cluster management requirements
  </Accordion>
</AccordionGroup>

## Scaling Strategies

### Vertical Scaling

Increasing single server performance:

* CPU and RAM increase
* Disk capacity increase
* Network bandwidth increase

**Usage:** When you want to use all resources

<Tip>
  After vertical scaling, update Gateway Worker pod thread and connection pool settings according to the new resources. For tier-based recommendations, see [Gateway Performance Tuning](/en/operations/administrator-guides/gateway-performance-tuning).
</Tip>

### Horizontal Scaling

Increasing server count:

* Increasing worker node count
* Increasing MongoDB replica set node count
* Increasing Elasticsearch cluster node count

**Usage:** For high availability and traffic increase requirements

<Warning>
  **Important Notes:**

  * **Do not use Test/PoC installations for load testing purposes!** If you want to evaluate the correct configuration for load testing, please refer to our [Benchmark Results](/en/concepts/benchmark-results) page or contact us.

  * **We do not recommend installing on a single server for production environments.** Please evaluate such an installation configuration only for PoC environments.

  * Requirements for each topology are for **minimum** configuration. They should be increased according to your service loads.
</Warning>