Market pressures are dictating that businesses move at increased velocity in order to respond faster to rapidly shifting customer demands, tastes and habits. Advancements in User Experience design, large scale computing, and advanced machine learning algorithms have expanded the reach of technology and conditioned customers to expect businesses and services to be easily reachable anytime and anywhere. These trends translate into requirements for IT organizations to build and sustain highly scalable, adaptive and resilient architectures that allow customers to be reached, served and understood on all platforms and through all channels quickly and efficiently.
To increase their agility, IT organizations are quickly adopting cloud-based architectures that offer the promise of distributed, highly scalable infrastructures coupled with drastically reduced procurement cycles. The real challenge, though, lies with the systems businesses rely on – ERPs, CRMs, Shopping Carts, and line-of-business applications that depend on well proven Relational Database Management Systems (RDBMS). RDBMS were not designed for dynamic and distributed cloud architectures and are proving to be the weakest link in application stacks. This is especially evident in hybrid cloud environments where latency between the nodes varies and connectivity is not always guaranteed.
IT organizations have tried to enhance the reliability of their RDBMS solutions in cloud environments by using legacy approaches like backups, log shipping, database mirroring, and database replication. In reality, they have only been frustrated by very slow recovery times, complex deployment and support needs, outdated legacy licensing models, and ultimately costs that far exceed the currently available budgets. The impasse has put businesses at risk of IT failures that in the era of the social network can easily turn into major brand damages with negative financial consequences.
What IT organizations are looking for is a solution that was designed to take advantage of cloud infrastructures rather than fight against them. They need a solution that will scale up or down as needed, be resilient to network performance disturbances, and provide self-healing capabilities in order to reduce support needs. What is needed is a platform to easily build Highly Available RDBMS deployments with heterogeneous nodes that may run different versions and sizes of the RDBMS in order to maximize the overall availability and minimize operating expenses.
