With a single managed server your applications will on run on just one machine. Positive's individual servers provide the greatest stability in the industry but all your eggs are in one basket. Even though we take consummate care of that “basket”, critical application require more than this. In the face of of adversity, even in the worst-case scenario, vital systems simply must remain available. This means going beyond the single-server solution.
A requirement of popular hosted application is that the system grow to keep up with demand, and that no “growing pains” accompany the additional capacity. Serious business applications require massive scalability for any demand, and they require this ramping-up of hosting to go smoothly, with absolutely no disruption to the application as a whole.
Business critical venture demand a cluster system. This spreads the load of the hosted application across any number of machines. For example, one machine might run a web-server and another handle the databases. As demand grows, further machines can be added to the database service, and spread the requests across these devices. You don't need to be an expert at load-balancing: Positive can design the system to fit your project and budget perfectly.
Furthermore we configure a cluster so that the failure of a single component cannot cause the whole system to die. For example, two database server machines might keep in perfect synchronisation, so that the failure of one is inconsequential to the availability of the database service as a whole.
With such a cluster solution, should a single machine fail, the system automatically moves over to use the hot-standby machine which had kept in perfect synchronisation. As with a multi-engined aircraft, the system as a whole does not crash, thanks to the usefully-redundant duplication built in.
Another benefit of the cluster system is in its increased security Different services lie behind different security zones, and are segregated so that a compromise is one part of the system does not affect security as a whole. Rather than trying to tune a machine to an average performance profile and security requirement, we can tune each aspect of a cluster to the specific task that runs of it. Such tailoring increases security, stability and throughput, and with Positive it comes as part of the standard Managed Cluster service.
Setting up a cluster system takes a large amount of expertise and experience. It can also take a large number of specialised components. At Positive, we have gained and unparalleled experience with technologies that minimise the cost and complexity of running such a system. These technologies, as ever are Open, completely flexible, and rock solid.
This is why banks, major broadcasters and international retailers rely on us to design, manage and run their hosting clusters: we provide a solution that just “works” and is tested in every conceivable failure-mode to prove its resilience. We manage the cluster as a whole, not just the individual machines, and leave you to do what you'd prefer to spend your time doing – running your business.
Positive Internet's resilient clusters have guaranteed reliability built from the ground up, including multiple power sources, network paths, power supply unity, physical cables and backups. For the ultimate in such guaranteed reliability, we offer geographical redundancy, where we can house a part of your cluster in one of our server facilities hundreds of miles from your primary system. This remote facility has its own, utterly independent networks paths to the internet.