This is our third post in the series that documents Maplewave’s journey to transition our IT infrastructure to the public cloud. Follow the links for Part 1 and Part 2.
———
In my last post, I discussed the financial advantages that Maplewave will gain by moving our hosted compute and storage environments to the public cloud. Based on our current and projected spend, we expect that we’ll see significant cost savings within 18 months while allowing us to increase availability, scalability and security.
In this article, I’ll dive deeper into the advantages that go beyond just the finances.
Forecasting Growth with Hardware
In our current hosting setup, we have typically oversized our hardware to allow for growth over time as most organizations are inclined to do. The tricky part is how much growth to expect and over how long?
This starts with multiple conversations with the executive team to understand the corporate growth strategy, more conversations with the sales and marketing teams to understand the sales funnel, and then making the best educated guess possible. But it’s still a guess. If we happen to land a large contract, it’s very possible that our current hardware would not be able to handle the additional load.
We would then be in a situation where we must purchase new hardware and/or additional rack space and bandwidth to accommodate the new business. Although it can be seen as a nice problem to have as we would have gained a nice new chunk of revenue, there is a true bottom line cost that we probably did not plan for.
At the same time, adding the additional hardware would greatly affect the timeline of the new project. Just procuring, receiving and racking the new gear is going to take weeks, even if it is readily available.
And what if the opposite were to happen and we then don’t need all the additional capacity? Now we have too much additional capacity that we still need to pay for, both in capital funding and operational expenses.
Forecasting Growth with The Cloud
So how does all this change when we look at a public cloud environment? First and foremost, it’s an on-demand world in the cloud. We only pay for what we use, and we only use what we need. Amazon Web Services are billed by the hour or the usage amount and can be changed virtually at a moments notice. So, suddenly, scaling up or down is not a limitation anymore.
Now that being said, we will definitely establish some service baselines that we expect to require over a 1-3-year period, so that we can take advantage of reserved instances at a better price point. Most of our customers are signed for a length of time contract, so linking reserved cloud resources at a lower cost totally makes sense.
Then, when we add a new large customer, we simply add additional resources as required. No more hardware procurement. No more hardware service contracts. And we can do it all in a matter of hours or less depending on the level of automation that we achieve (hmmm, I sense another blog post coming).
Our ability to scale and grow can now be matched with our ability to sell our services in the global market and meet or exceed our corporate growth strategy.
Availability
Speaking of global markets, this provides a beautiful segue into the other major advantage that we will derive from the public cloud: availability.
Availability means that our services are there for our customers to use when they need them. Amazon’s EC2/EBS SLA provides an uptime of 99.99% when services are spread across multiple availability zones within a single region. As a service provider to our customers, this allows for a more than adequate starting point to then layer on our application availability and still come out with some very good uptime numbers.
But availability is more than just uptime; availability means that no matter where our customers reside, our applications should be available for them to use with the same level of performance that they would achieve if they were local to our data centers. With eighteen regions and fifty-five availability zones (and more planned), AWS is pretty much where our customers are located today and in the future.
Being able to market ourselves as capable of deploying services in Australia or Europe without any latency issues opens a world of possibilities for a company based in Halifax, Nova Scotia.
In the end, financial savings, although significant, pale in comparison to the business benefits Maplewave will derive from increased scalability and availability. Growth and innovation drive our business and the public cloud will be the engine.
Next Time
In my next blog, I’ll discuss cloud security and why I believe we’ll be even more secure after the move to the public cloud.
