By Sean Rinas, Head of Network Operations
If network outages and their downstream impact on your dispersed IT infrastructure keeps you awake at night, you’re not alone. Network-related business continuity issues are on the rise. Recently released Uptime Institute research placed network-related outages (at 29%) as the No.1 cause of serious IT downtime.
This number alone is case in point as to the criticality of establishing an interconnection-first architecture that proactively seeks out any potential point of failure.
It was 16th century Dutch philosopher Desiderius Erasmus, who first said “prevention is better than cure”. It’s a central tenet of strategic risk management that applies equally to network resilience as it does to a health and wellbeing or any other strategic planning.
In the digital realm, complexity is a constant and interconnection is the lifeline to cost, productivity and risk mitigation benefits; so, you need good preventative discipline to avoid painful cures.
Connectivity architectures that are mapped, measured and managed to achieve the same fault tolerance goals as infrastructure platforms are the only ones equipped to deliver best availability outcomes.
Unsurprisingly, Uptime’s research found 43% of respondents who had experienced a network outage in the last three years, could attribute it to a “third-party network provider failure”.
Uptime warned that networks are both technically and operationally complex. It also pointed out that “ownership, visibility and accountability can also be complicated” all of which introduces the possibility of blind “overlapping points of failure or capacity pinch points”.
Where complex and high throughput environments predominate, even seemingly insignificant failures or configuration errors can cascade across networks leaving no simple markers to conduct root cause analysis.
Vigilance is essential when it comes to building redundancy with 100% geodiversity in fibre routes, switches and facilities. Third-party services rarely offer the controls to manage this. Interestingly, the same Uptime research found that those who had avoided outages in three years, paid the most attention to controllable factors such as diversity and redundancy.
Hybrid IT is now essentially ubiquitous across the economy. Despite its pros, there are also cons of meshed physical and virtual infrastructure in the form of complexity, resulting in customers needing more real-time visibility and control over their services and routing within their own domains.
For cloud to-cloud connectivity, Layer 2 services are becoming the preferred option. Not only because they allow organisations to tightly manage their security and resilience posture, it also enables self-provisioning capability of private connections at the performance and duration required under a PAYG arrangement.
As an example, most industries operate under strict compliance obligations to retain communications, transactional and other data for extended periods, sometimes for as long as 7-10 years. Databases are updated, backed up and archived continuously, and where critical or sensitive data is involved, sometimes duplicated in a tertiary environment for resilience assurance.
Where data archiving for compliance is concerned, cloud is often both the source of data and the destination for its archiving/storage. These are largely mission critical transfers that often involve very large datasets that have traditionally been dependent upon complex service agreements which interconnect multiple distributed compute locations.
For example, an organisation may wish to archive cloud-native Outlook, Salesforce or ServiceNow data using one of the public cloud platforms for compliance storage.
Finding efficiency and cost gains here without compromising security and resilience imperatives is a huge gain for organisations which allows them to claw back resources and divert them to other prioritised agendas such as innovation and growth.
Meanwhile, in industries such as media & entertainment or mining & exploration, engagement with cloud platforms is core to content delivery innovation initiatives or as secondary infrastructure managing ML and AI workloads.
Similarly, in mature organisations, there could be hundreds of similar large data transfers that take place periodically, or as a one-off back-up, innovation projects and big data analysis being shared around people, locations, clouds and service providers.
Regardless of the use case, these transfers require a flexible, resilient and secure point-to-point service at a prescribed capacity for a period that’s determined by the business. Increasingly, organisations are opting for intuitive interfaces offering self-service provisioning and complete visibility of routes and performance.
This is where flexible, intelligent networks that can identify congestion or outages as (or before) they happen and automatically re-route where required ensuring there is no service interruption. This self-healing capability ensures performance is always optimised.
When it comes to Hybrid IT, organisations are increasingly recognising the value of colocating their physical IT footprint adjacent to their digital services ecosystem. In doing so it enables IT leaders to address the criticality of security and protecting their data, in rest and in flight via an interconnection-first strategy.
Not only are organisations benefiting from near-zero latency for performance-sensitive interactions, they’re also successfully advancing security outcomes from data transmission via private connections and data hosted within a sovereign ecosystem.
Reach out to NEXTDC to better understand how we can help you build network resilience and redundancy into your interconnection strategy.