Now comes Software Defined Availability

No sooner has Cloud Services World predicted that `software enabled everything’ will be one of the hot topics for 2014 than a new software-enabled something appears.

This one comes from Stratus Technologies, which has announced what it claims is the first Software Defined Availability Platform (SDAP). Known as everRun Enterprise, it delivers mainframe-like reliability to virtualised workloads and includes such new features as disaster recovery and application monitoring to ensure business continuity.

“This launch signals a change in strategy for Stratus,” said Nigel Dessau, Chief Marketing Officer at Stratus Technologies. “We are broadening our focus on fault tolerance – which ensures the highest levels of availability – beyond hardware, to virtual workloads on commodity servers. In effect, we are bringing mainframe-like levels of availability to Intel-based servers.”

Scheduled to be available at the end of February, everRun Enterprise unifies the ease of use, monitoring, management and service capabilities of the Avance high-availability software with the fault-tolerant engine of previous generations of everRun. Additionally, the entire solution now leverages the open source Kernel Virtual Machine (KVM) hypervisor. The result of this new architecture is a higher performance software platform that supports a wider range of operating system guest workloads.

The company is targeting businesses that have so far resisted migrating business-critical applications to virtualised architectures due to a lack of reliable, enterprise-grade availability in today’s hypervisors. The company also sees a very similar requirement in private clouds, so the new everRun technology will be the basis of the company’s cloud strategy as well.

The goal is to move the capabilities of downtime prevention and recovery from hardware to software, and Stratus is claiming it is the first company to deliver a solution for an emerging category that it calls Software Defined Availability.

Examples of the problems that can come with unexpected downtime amongst businesses and service providers have hit the headlines a few times of late, and according to the Aberdeen Group, the cost of downtime increased by 40 percent in the three year period 2010 to 2013.

everRun Enterprise is designed to keep Windows and Linux applications up and running continuously without changes to applications or the occurrence of any in-flight data loss. It is compatible with a wide range of platforms, is easy to implement and includes centralised management tools for an all-encompassing view of the entire stack – from bare metal to the applications.

The software runs on industry-standard Intel-based x86 servers without the need for specialised IT skills. Unlike other availability solutions, everRun Enterprise prevents downtime from occurring rather than merely recovering from it – a difference that the company suggests can have a significant impact on revenues, costs and customer satisfaction.

Basis of the system is the Availability Engine, which achieves always-on availability with applications mirrored in two virtual machines with no data loss or machine restarts. This can work as what the company calls a Split-site Cross Campus system, which provides application fault tolerance across geographically separated sites.

It provides Disaster Recoveryby using a built-in asynchronous replication between sites over a wide area network connection, and management is provided by a Stratus One View console which enables users to build, designate, deploy, monitor and manage everRun instances and virtual machines from one centralised location.

There is a System Watchdog and Alerting Service that constantly monitors systems and automatically sends a system-level notification should a fault occur, as well as Application Monitoring to reduce downtime. This can simplify, accelerate and optimise performance by providing visibility across IT environments and delivering automatic fault diagnosis.