Stratus has launched Avance which combines the benefits of high-availability and virtualisation into a single integrated solution.
The software is able to identify, isolate and report faults while full resource sharing enables the system to use unaffected components if a hardware or software error is imminent. It is designed to enable users to replace disparate stand-alone servers with a single solution that is easy to deploy and manage.
IT professionals can remotely monitor and manage the virtual machines, the physical x86 systems and network interfaces. The software immediately begins monitoring the operation of your physical servers as they work and their virtual machines can predict events and handle faults — without intervention by a skilled operator.
Applications ride through most causes of system interruption, including disk and network subsystem failure. As a result, you can avoid the processing blackout, data loss and user disconnects that happen during a conventional cluster failover.
Avance software automates fault detection and error handling and continually monitors the heartbeat of virtual machines. Upon sensing a problem in one server, the software migrates applications to the unaffected server while the problem is isolated to the component level for rapid repair. Software and hardware faults and other events are automatically reported via email and custom reports to your IT staff and/or another authorised service provider.
Avance High Availability also helps eliminate planned downtime for hardware and software upgrades, as by using the management portal to transfer production to the other member of your server pair, your virtual machines won't miss a beat. When you're ready to bring the upgraded system back online, Avance automatically resynchronises the application(s) and disk mirroring.
Andy Bailey, availability consultant at Stratus Technologies, said: “This is specifically to tackle high availability systems, where there is no user intervention unless you have the concept of continuous availability.
“There is a gap in the market where people want high availability but cannot afford the server, they want industry standard systems but they are in edge type environments where IP professionals are few and far between.
“This works by diverting problems from one node to another if there is a problem, if one node fails we have the concept of moving servers to a healthier node which is why we say we can cover most hardware faults – if the primary node has a problem it will re-start. There is always a chance the automater will do something wrong so it is automated to override problems.”