- Cluster History
- Clustering for Fault Tolerance
- Distributed Systems Clusters
- Failover Clusters
- Load-Balanced Clusters
- Conclusion
Like this article? We recommend
Like this article? We recommend
Like this article? We recommend
Distributed Systems Clusters
A distributed systems cluster is a group of machines that are virtually or geographically separated and that work together to provide the same service or application to clients. It is possible that many of the services you run in your network today are part of a distributed systems cluster. Domain Naming System (DNS), Windows Internet Naming Service (WINS), and Active Directory are all examples of services that are distributed; yet provide a service to users. So what makes it a cluster? It is defined as a cluster because the servers are fault-tolerant, and they provide seamless access to a service. Even though each server provides a service to the clients that are on the local network, any server within the network can respond to requests. This is further defined as a cluster because:
- There is no single point of failure (SPOF).
- The design provides scalability.
- The data is shared between all machines.
- The design is reliable.
The distributed systems cluster eliminates the SPOF by providing redundancy in the data access. No SPOF will cause the service(s) to be unavailable. It provides scalability because more servers can be added to each location to field client requests, and new sites can be added to the WAN without necessitating an upgrade of existing systems. The data is shared between all machines, making the data accurate to clients that request it. Overall, the system is reliable and provides service seamlessly as one virtual server, even though many servers are involved.
Distributed systems clusters work well in corporate LANs and WANs, but they don't fit every need. For Internet applications, connectivity takes place through one connection — an Internet connection — and therefore is not the best solution for systems where the Internet contains the client base. For other applications, where the data transfer is minimal or the protocol is lightweight, a centralized system may be more desirable.