Infrastructure is still critical
With the constant clamour about new technologies, digital disruption, and transformation, it’s easy to get caught up and carried away in the hype. In doing so we risk giving less attention to business continuity and service availability provided by infrastructure and platform critical to keeping a business operating.
Complacency in this regard can lead to devastating results, especially in a world where social media quickly exposes even the slightest failures of service, let alone how they are dealt with. Core infrastructure that is designed, built and operated with the sentiment of “planning for failure” at heart is vital to ensuring a business can operate in situations not desired for.
Recently, I’ve been affected by an infrastructure failure when the lovely tree lined streets of my suburb came tumbling down and took all the overhead powerlines out with them. Fortunately for us this failure wasn’t life threatening, but we know that for some it was. We did suffer our own crisis as the household was faced with the proposition of no power to charge the iPhone, and in finding matches to light the gas hob (we are fortunate in that we had gas). The prolonged lack of power caused us to do what everyone else did, we made the most of a tough situation, we leant on friends to help, we provided what we could to the household, but it was certainly a reduced set of services.
In business we rely on our infrastructure always being available to provide service to staff and clients. In the event of a failure it’s not simply a matter of popping down the road for a shower and to borrow some power. It’s critical that we can continue to provide services to our clients and to operate the business and an outage of any substantial nature can be damaging to the brand, can be costly, and can result in losing business and clients. While it may not be fashionable to talk about technology infrastructure it is still vital to the effective operations of a business, and as such, care needs to be applied to how we design, build, and operate it.
With the continued development and availability of new infrastructure services, and the perceived ease of migration and use, it’s easy to become complacent about what are in fact critical infrastructure services. The adage of “close enough is good enough” may suit some circumstances, but it never does for critical services that affect the health of the business, or the wellbeing of people. If you’re reliant on a service, and it goes away simply due to poor planning, how will that effect you, and does it actually matter if the service is free? When planning infrastructure to support the business and its services you (the business) need to take responsibility for ensuring that it is designed well, it’s secure, stable and resilient. It’s important to do this for both physical and virtual infrastructure services, consuming an aaS offering from a third party does not exempt you from the responsibility of service assurance, even if you can blame someone else for a failure.
One of the most important concepts you can use in planning infrastructure is to “plan for failure”. Colonel Chris Hadfield talked to this concept in the recent IBM Think conference in Auckland, and he’s well known for promoting this concept. The concept isn’t revolutionary at all and if you google “plan for failure” you’ll see a wide variety of use cases. When planning for and designing infrastructure it is vital that you bring this concept to the forefront of the discussion, we used to think of this as the five 9’s availability but we rarely see that term related nowadays. Even when we do see attention given to availability and uptime it’s not always synonymous with testing failure scenarios and planning for failure. Make it part of your operational process to seek out new scenarios for failure.
We’ve been designing, building, and operating infrastructure/platform services in NZ for over twenty years and in that time, we’ve seen numerous waves of technology reshaping the way we and our clients work. The fundamentals of providing business platforms are largely unchanged and it’s important to reflect on and review these things often:
- Infrastructure and platform services may not be fashionable, but they are critical, treat them as such.
- Don’t be complacent in the lifecycle of your infrastructure (even if it is aaS), it needs to continue to be secure, stable and resilient.
- Plan for failure and test regularly.