High Availability VM

Even for mission-critical web sites and applications, where 99.9% uptime simply isn't good enough, Net Logistics remains the preeminent choice in the Australian hosting industry. For those mission-critical applications that require significantly higher uptime, we are pleased to not only offer a full spectrum of High Availability (HA) hosting platforms, but also provide these HA services across a very wide and flexible feature spectrum (ranging from shared hosting to full virtual dedicated server solutions). Each of these platforms is specially-engineered to eliminate any single point of failure, with automatic failover in the event of component or sub-system failure. By extending this concept to multiple layers — e.g. redundant front-end machines, redundant hard disks via RAID and dual SAN, redundant network equipment, and redundant network connections — and combining this with high-priority support, aggressive proactive monitoring, and system engineering practices such as clustering, we can achieve extraordinary availability, and still maintain surprising affordability.

Each of our platforms benefits from an exceptional infrastructure, resulting in uptimes exceeding 99.98%.

Infrastructure Overview

As mentioned, we have ensured that the entire architecture has no single point of failure. This includes our core network, storage, and blade center hardware itself. Other than physical server resources, our entire HA architecture is configured with n+1 redundancy. To protect against server failure situations, we conduct automated off-site backups for all HA platforms. The net result is rather impressive. For example, in the event of a network switch failure, load is automatically transferred to a secondary switch within five seconds. In the event of hardware failure, complete system load is also automatically transferred to another server — what would typically take hours and require constant technician involvement, is automatically accomplished within the time of a single reboot (typically only a couple of minutes). Complete fault-tolerance is also available as an additional feature on some of our platforms, enabling load to automatically be transferred to secondary virtual machines in less than two seconds. Even more extreme events, such as complete storage failure or catastrophic data center events (including complete destruction of the datacenter) are fully planned for, and can be rapidly resolved. Following is a brief overview of our HA individual service offerings.

You can read more information on each of these platforms on their respective pages, or contact us with any questions you may have.