Server Load balancer is a type of cluster deployment. It uses multiple nodes to perform the mail service. Each new request to obtain the web page is dynamically routed to a node with lower load. That is, the load pressure is allocated to each computer in the cluster according to a certain algorithm to reduce the pressure on the master server and reduce the hardware and software requirements on the master server, which is suitable for large-capacity users.
Cluster Technical Analysis
Cluster technology refers to a group of independent computers that use high-speed communication networks to form a computer system. Each cluster node (that is, each computer in the cluster) is an independent server that runs its own processes. These processes can communicate with each other. For network clients, it is like forming a single system. They work together to provide applications, system resources, and data to users and manage them in a single system mode. When a client interacts with a cluster, the cluster is like an independent server.
Cluster Features
- Provides high-performance computer systems with powerful processing capabilities. For a client, a computer cluster is a single system that can provide users with a high-performance computer system. users do not have to worry about how many computers are responsible for system implementation, you only need to pay attention to the overall processing capability of the system. Therefore, a computer cluster can use multiple computers with common performance to form a high-performance computer system. Only the supercomputer can be competent.
- Provides high-availability computer systems. A system composed of Computer Cluster Technology ensures the high availability of data and applications to end users, and manages the types of faults. That is, when a node or computer in a computer cluster experiences a hardware or software fault, the high availability cluster provides replacement for the software and hardware failure. It mirrors the server to a backup system or node. When the system on the master node crashes, the redundant node switches from the backup role to the formal role and is automatically applied to the application, this ensures uninterrupted system operation.
- The system has high scalability: when the system's processing capability needs to be increased, in addition to increasing the single-host processing capability of each computer node in the cluster (such as increasing the number of CPUs and memory size), you can also increase the number of cluster nodes, that is, by adding new computer nodes to the cluster, services can be scaled as the processor is added, increasing the application throughput to increase the overall processing capability of the system, complete System resizing.
Problems solved by Server Load balancer
Based on the existing network structure, each service node in the server group runs an independent email program. Server Load balancer provides a cheap and effective way to expand server bandwidth and increase throughput, strengthen Network Data processing capabilities to improve network flexibility and availability. It mainly performs the following tasks:
- Solves network congestion and provides nearby services to achieve location independence.
- Providing users with better access quality
- Increase Server Response Speed
- Improves utilization efficiency of servers and other resources
- Prevents single point of failure in key parts of the network
Network Topology
- Server: 2 servers. Install all turbomail modules.
- Network: Use networks of all levels according to the specific needs of the Intranet and Internet.
- Storage device: a disk array is used to store turbomail data, including user and email data. storage devices such as NAS and San are also used.
Server Load balancer includes two servers with shared data storage space. Each Server communicates with each other through an internal LAN. This solution has several advantages:
- High reliability: When a node fails, the applications it runs will be automatically taken over by another node to provide uninterrupted services to users.
- High-performance computing: Make full use of the resources of every computer in the solution to achieve parallel processing of complex operations.
- Load Balancing: distributes the load pressure to each computer in the cluster according to a certain algorithm to reduce the pressure on the master server and reduce the hardware and software requirements on the master server.
- Unified mail data management and maintenance.
- Turbomail bandwidth access solution.
Two-line access is recommended for North-South communication. The bandwidth access calculation method is as follows:
Turbomail mail 10 thousand user configuration instructions:
(Calculated based on 20 k mails sent and received by each user per day)
A. SMTP/POP3 request:
100 K (email) * 20 mails * 10000 (User) = 20 Gbyte = 160 Gbit/86400 (24 hours * 60 minutes * 60 seconds) = 1.85 Mbit/0.6 (Ethernet bandwidth utilization) = 3.083mbit * 4 (peak bandwidth ratio) = 12.35 m.
B. Web mail request:
100 K * 20 mails * (page requests) * 3000 users (3/10 of actual Web Access ratio) = 6 Gbyte = 48 Gbit/86400 (24 hours * 60 minutes * 60 seconds) = 0.555 Mbit/0.6 (Ethernet bandwidth utilization) = 0.8335 M * 4 (peak bandwidth ratio) = 3.35 m.
C. Suggestions:
The Internet access bandwidth of 10000 users is 12.35 m + 3.35 m = 15.7 m -- 16 m.
