Zenlayer Load Balancer Overview

Zenlayer Load Balancer is a next-generation Layer 4 load balancing service launched by Zenlayer. It delivers ultra-high performance and automatic elasticity, with a single instance supporting up to 5 million concurrent connections, enabling you to easily handle high-concurrency workloads.

Product Advantages

Ultra-High Performance

Load balancer uses a self-developed user-space forwarding engine to achieve high-performance packet forwarding. Its device clusters can scale horizontally, with each single instance supporting up to 5 million concurrent connections and up to 20 Gbps of throughput, making it well suited for high-concurrency scenarios with massive numbers of connections.

Automatic Elastic Scaling

You do not need to specify or manually adjust load balancer instance specifications. Instance capacity automatically scales up or down in response to changes in your business traffic.

High Availability

Load balancer adopts a multi-layer disaster recovery architecture to ensure service availability, including cluster-level redundancy, session persistence, and independently deployed health checks. The load balancer control plane periodically monitors the operational status of forwarding-plane members. If an abnormal node is detected, it is removed from the cluster and automatically rejoined once it recovers. Independently deployed health check nodes perform periodic health probes on backend servers based on the load balancer instance configuration. If a backend server is found to be unhealthy, its status is marked as unhealthy and the load balancer cluster stops scheduling new requests to it. Once the server recovers, its status is restored to healthy and it resumes sharing the traffic load.

Multi-Scenario Traffic Distribution

Load balancer supports attaching IP-based backend services. When integrated with products such as Border Gateway, it enables traffic distribution and scheduling across multiple scenarios, including cloud environments and on-premises IDC (Internet Data Center) servers.

---

Application Scenarios

Internet-Facing Cloud Service Entry Point

As the entry point for internet traffic, load balancer provides ultra-high Layer 4 processing capacity with a single instance and automatically scales based on business demand. No manual intervention is required during traffic fluctuations, significantly reducing O&M overhead.

IoT Service Entry Point

In scenarios such as smart homes, smart parking, video surveillance, and vehicle networking, load balancer serves as the service entry point to handle massive concurrent connections while providing security group capabilities to ensure the secure and stable operation of IoT workloads.

Hybrid Cloud Service Entry Point

Load balancer supports attaching on-premises IDC (Internet Data Center) servers. When combined with products such as Border Gateway, it can forward cloud-based requests to on-premises servers, enabling seamless connectivity between on-premises IDC environments and cloud services.

---

Key Features

Zenlayer load balancer is a network load balancer. Its key features include:

NAT Mode

No additional configuration is required on backend servers. Backend servers can operate normally using only private IP addresses. The load balancer rewrites the destination IP address of incoming requests to the selected backend server’s IP address while keeping the source IP unchanged. As a result, the backend server receives the real client IP address.

TCP Listener

Load balancer forwards TCP protocol requests from clients. This is suitable for scenarios that emphasize reliability and data integrity, where speed can be relatively less critical, such as file transfers, email sending/receiving, and remote login.

UDP Listener

Load balancer forwards UDP protocol requests from clients. This is ideal for scenarios that prioritize real-time performance over reliability, such as video calls and real-time financial market data delivery. You can add a UDP listener to a load balancer instance.

Maglev Hash Scheduling Algorithm

Maglev Hash is a consistent hashing–based load balancing algorithm proposed by Google. It provides uniform traffic distribution, high performance, and excellent scalability.

Round-Robin Scheduling Algorithm

Requests are distributed sequentially to each backend server in turn. This algorithm is suitable when backend servers have similar performance characteristics.

Health Checks

Load balancer uses health checks to determine the availability of backend services. When health checks are enabled, if a backend server becomes unhealthy, the load balancer automatically routes new requests to other healthy backend servers. Once the server recovers, it is automatically reinstated and resumes handling traffic.

Security Groups

Security groups enhance the security of cloud traffic entry points. If you need protocol-, port-, or IP-based access control for load balancer listeners, you can achieve this by configuring security groups for the load balancer instance.