Pay-Per-Minute Billing Models
Authored by Volker Rath, Hosting Subject Matter Expert
Why can’t I buy cloud computing at Macquarie Telecom by the minute, so I can just pay for what I consume?
Pay by the minute billing models are touted by some competitors as a cost-effective option, but they are usually not of good value. Competitors actually educate the market by proclaiming that this is the best way to go. We strongly disagree. If you require hosting or a provider for business-critical infrastructure, you are not setting a system up for only a few minutes of use. You do not need an email server for five minutes a day, and you do not need a web server for only three hours a day. Let’s investigate why some companies can charge by the minute while others do not.
First of all, it all comes down to the solution stack. This is not only a commercial decision made by an individual provider. If a solution stack is built upon standard building blocks that allow for zero flexibility, automated solutions can be fully provisioned and decommissioned in a very short time frame. However, if a company wants to host business-critical infrastructure, this requires a bespoke solution. A special firewall unit and dedicated load balancer must probably be deployed. The infrastructure is running on a hybrid environment, such as a mixture of colocated dedicated servers and cloud servers working together.
This can be analogised as the difference between riding a bus and driving a rental car. The public cloud built from standard components allows for a pay-per-minute or pay-per-hour billing model similar to the public bus. Sometimes, the bus is full and customers cannot get a seat. Even when customers do get a seat, they still have to ride along the bus route, no matter how inconvenient.
If travellers needed more flexibility, they would obtain a rental car. This would allow them to choose a car that is customised to fit their needs (e.g., different sizes, GPS options, etc.). Additionally, the rental car company would allow its customers to drive wherever they wish. Of course, it is not possible to obtain a rental car for only a few minutes of use. The model would not work by its inherent design. Most businesses would not want to allow their business-critical infrastructure to rely on the bus-like pay-per-minute business model. They need bespoke, individualised solutions that are well-supported.
Business-critical infrastructure cannot afford downtime.
Pay as you go models have different kinds of service level agreements. While SLAs should not be the infrastructure’s life insurance in case the system were to go down, a major focus must be paid to uptime. We are now living in a society that is always connected. Everyone is constantly on the internet with their smartphones and computers (even your fridge or toaster might be connected devices at this point). Therefore, it is absolutely critical that organisations match the always-connected nature of the internet to provide 100% uptime.
Riding a bus just isn’t good enough for a robust solution. You would end up depending on too many other organisations. Uptime guarantees require guaranteed environments. Connectivity and resource availability should not be impacted by neighbours using the same public environment. That means that your systems should be run on a dedicated environment, minimising shared components as much as possible. These bespoke environments are harder to build and support. That is why it is critical to find a partner that can deliver on performance and uptime, along with guaranteeing the service you need for your business.
If you would like to ditch the public bus, get in touch so we can speak further.