Linux Clustering

Attention: The information presented on this page is out-of-date. In order to get the most actual info on StarWind products, please visit the following page.

The content here seems to be out-of-date. To get the latest info, please visit the following page

Linux Clustering

The term "cluster" generally means a group of computers joined together to generate more computer power or to make it possible to achieve higher availability (HA).
Currently, Linux clustering has become widely used in various industries. Because of the acceptance of clustering technology and the increasing popularity of open source software, companies can create supercomputers for costs that are much lower than the expenditure for high performance machines. Today, this is why it is popular to utilize commodity workstations operating under Linux or other open source software as a node in a cluster.

In general, there are three kinds of clusters:

Fail-over clusters
Load-balancing clusters
High-performance clusters

Each of these types may be highly useful for particular purposes and they are definitely worth being described below.

Fail-over clusters

A common fail-over cluster consists of two nodes. One node is active while the other node stays in a stand-by mode although continuously checking if the active one works properly. If the active node fails, the stand-by node takes over the workload and enables uninterrupted operation of mission-critical applications and availability of important data.

Load-balancing clusters

Load-balancing clusters are usually utilized for continuous operation of the most visited websites where more than one node is hosted. In this case, every incoming request for a web page is directed to a less loaded node.

High-performance clusters

This type of cluster allows simultaneous operation of programs for time-consuming calculations and is commonly utilized by scientific institutions. Such clusters often use simulations and other software that requires powerful CPU and large memory volumes. In these cases, high-performance clusters can complete tasks that would otherwise require extensive time.
With Linux and other free open source software, it is possible to create robust clusters at affordable prices. According to production environment requirements, more nodes can be added to a cluster anytime. Also, Linux allows its users to adjust the kernel to their workloads. It is considered to be one of the best platforms for building of high-performance clusters.

Linux HA is a software solution that provides availability and reliability. Its portable cluster manager, Heartbeat, allows building an HA cluster and has the following features:

Heartbeat: Enables creation of both large and simple clusters and allows adding as many nodes as needed since there is no fixed limit in the number of nodes.
Resource monitoring provides automatic restart of resources in case of a failure.
Fencing mechanism removes the nodes that go down from the cluster.
Sophisticated policy-based resource management, resource inter-dependencies, and constraints.
Time-based rules allowed for different policies.
Several resource scripts (for Apache, DB2, Oracle, PostgreSQL etc.) included.
GUI for configuring, controlling, and monitoring resources and nodes.

Learn more about Clustering:

{}