Load Balancer

Distribute workload across groups of backend resources to provide High Availability and improve your application performance.

Network Load Balancer (NLB) helps distribute your incoming traffic across groups of backend resources and also multiple targets such as NCS instance and container. Our Load Balancer operates at layer 4 of the Open System Interconnection (OSI) model, which supports these two use cases:

  • External Load Balancer: provides High Availability and helps expose your clustered application to the internet by using a single public IP. There is no need to use multiple public IPs, which leads to higher prices.

  • Internal Load Balancer: helps distribute traffic inside a virtual network and provides High Availability to your application or service.


High Availability (HA)

Our load balancer has 2 instances (VM) operating in active-standby mode to provide High Availability (HA). Each instance runs on different compute nodes to prevent downtime from any failures.


Multiple Availability Zones

There are 2 Availability Zones available, Bangkok and Nonthaburi. You can set up a load balancer on any AZ, depending on what your applications need.



You can manage the load balancer by yourself through NIPA Cloud Space portal. There is no need to manually provision HAProxy/Nginx. This is simple, time-saving, and suitable for the applications or services that require the use of load balancer.


TCP/UDP Forwarding

Support requests from layer-4 protocol.

Weight Configuration

The weight of an instance determines the portion of requests or connections of the services compared to the other instances of the backend group. A valid value is from 0 to 256. The default is 1.

Health Check

The backend group is always monitored. If an instance is not healthy, the load balancer will stop distributing traffic to that instance.


You can choose one of these following load balancing algorithms for your backend group.

  • Least connections: requests will be sent to the server with the least number of active connections.

  • Round robin: requests will be sent to each server in turn.

  • Source IP: provides stickiness between a user and a server. A request will be sent to the same server previously used.

Performance Types

We provide 2 performance types of the load balancer to select what best suits your applications.

See Pricing
Standard Load Balancer

Standard Load Balancer

Resources are shared between load balancers, so you can use the load balancer at a lower cost.

Dedicated Load Balancer

Dedicated Load Balancer

The dedicated load balancer enables exclusive use of resources, resulting in superior performance that is not affected by others.

Resources are shared between load balancers, so you can use the load balancer at a lower cost.

This standard load balancer allows your team to use up to:


Active Connections

2 Gbps


Upcoming Features

  • Application Load Balancer (HTTP/HTTPS) layer 7*

  • Content-based Routing (Layer 7 Policy, Layer 7 Rule and Layer 7 switching)

  • HTTP and TLS-terminated HTTPS Load Balancing on the same IP and backend

  • TLS termination/SSL termination

    • Server Name Indication (SNI)

    • Client Authentication (two-way authentication)

    • Secure HTTP/2 with Application-Layer Protocol Negotiation (ALPN) TLS extension

*Reference from the OSI model.

Want to try?

Get Your First Month FREE Trial

with Migration Support