CHASSAsia, by using a simple and innovative technology, is able to deliver a higher level of Windows application availability at a lower total system cost than clusters.

Cluster Configurations
Clusters are groups of servers configured to increase application availability. Cluster management software on each server monitors the functioning of every server in the cluster. Typically, clusters have shared storage to assure data integrity. If the software detects a failure, it begins a process called a “failover” that transfers the workload of that server to another functional server in the cluster.

Clusters have three key weaknesses which should be considered before choosing an availability technology: complexity, downtime risk, and cost:

	1. Failovers Mean Added Complexity To failover, applications must be written and configured to automatically restart on a another server after a failover. This scripting and the extensive testing required by the failover process represents an added burden to IT resources.
	2. Risk of Application Downtime Clusters do not provide continuous application availability. Even during a successful failover, the application stops working for several secords (sometimes for several minutes.) In addition, the complex failover process often does not execute properly, making the application unavailable for hours. Adding to this downtime, applications are also taken offline during the “fail back” process to restore a protected configuration. Another source of downtime risk is the single shared “quorum disk” (often shared over a costly fiber channel connections) that maintains cluster state. If it fails or becomes full, the cluster will not function.
	3. Clusters Add Cost Clusters add significant cost to the applications they protect. Most applications are “non-cluster aware” or significantly more expensive than standard versions. Some clusters require expensive additional software to monitor application operations and manage failovers.

Marathon FTvirtual Server
Marathon FTvirtual Server software uses an innovative technology to protect applications from downtime while eliminating the cost and complexity of clusters. Marathon software creates a virtual application environment that appears on the network as a standard Windows host server. Any standard Windows application can be installed, configured, and run in the same way that they are on a standard server. Marathon software enables the virtual Windows environment to run on two servers simultaneously by keeping the servers in synchronized “lockstep.” All application operations are performed on both servers at the same time. If one server fails, the virtual Windows environment continues to operate without requiring a failover process using the computing power of the functioning server.

Simple Administration Saves Money
Marathon software protects any standard Windows application without the modification or scripting. Servers can be repaired and replaced without special testing and while the application is continuously online.

Continuous Availability
There are no single points of failure in a Marathon configuration. Component failures in either server (and even certain combinations of failures in both servers) and software errors such as faulty device drivers do not interrupt processing.

Marathon configuration can be repaired while the application is continuously online. The failed server is taken off line, repaired, and rebooted. Both data and application are automatically copied from the functioning server to the repaired server as a background task. The application never pauses, loses state, or drops data. Data in the Marathon FTvirtual Server are continually mirrored between the two physical servers, eliminating the added expense of shared disk arrays.

Marathon FTvirtual Server software saves money, protects data, and delivers simple, affordable, continuous uptime for all your Windows applications.

          Click here for Marathon Success Stories - (Leading sports & Media company in US)