We’ve come a long way since the birth of Web In A Box. Our infrastructure has grown and evolved with thoughtful decision making and structured planning.
Our first priority has always been to deliver the best service possible, so we’ve designed our hardware to ensure we can provide the fastest, most reliable hosting around.
Our network infrastructure was designed from the ground up to have multiple points of redundancy. It features multiple upstream transit providers, a load balanced and clustered server architecture, redundant storage and enterprise internal and external monitoring.
We are multi-homed with multiple upstream transit providers, and connectivity to all major peering points in Australia. Our location within NextDC P2 in Perth gives us physical proximity and connection to transit and peering partners reducing a point of failure in metro fibre services. We support both IPv4 and dual-stack IPv6 and connectivity is provided by redundant transit providers, as well as multiple 10Gbit peering links to WAIX and EdgeIX Perth
Our hosting clusters are built on a combination of Ubuntu GNU/Linux and Windows Server operating systems deployed on the latest high-end AMD HEDT and Server platforms, while our storage servers utilise multi-path 10 gigabit with high speed flash storage.
Our Virtual Servers use industry leading VMware hypervisor technology and a Xen based platform for the highest availability and performance.
Finally, our network uses Cisco routers, Juniper and Cisco Catalyst switches, delivering unparalleled performance and reliability.
Our figures really do the talking for us – our network uptime for the last year exceeded 99.99%.
While our network is built to minimise the impact of possible issues on our clients – we recognise that problems can and do occur. When something goes wrong, we aim to fix it fast, let you know what happened, and ensure it doesn’t happen again.
For planned outages (such as maintenance and testing) – customers will be notified in advance to facilitate planning, and we will schedule an outage at a time which minimises any impact on our customers.
With internal unplanned outages (such as internal network issues, server issues) – we have a 1 hour response time to the customer from time of notification of the issue, to advise of an estimated time of restoration (ETR). In most cases, it is unlikely that an ETR will exceed 3-4 hours barring critical failure.
For external unplanned outages (such as upstream provider issues) – we invoke our SLA with our vendors, and will notify a customer within 1 hour of an ETR pending availability of vendor support. Where possible, customers will be notified with updates as they become available every 2 hours.
Read more… Business Continuity (PDF)
