We've all gotten in way over our heads at some point or other. Perhaps it was at college when your schoolwork, social life and hopeless laundry situation made you feel overwhelmed. Or at home when your kids, spouse, and dog all needed you at the exact same moment. Or perhaps it was at work when your AWS project turned into a runaway success but the size of hosting bill threatened to bring the whole thing down - is it even possible to optimize AWS costs?
Success at work is great and a successful cloud migration is very gratifying. But costs can, and often do, spiral out of control. When it comes to managing your AWS infrastructure, Reserved Instances (RI) may be the lifeline that you're looking for.
What are RIs? Let's break it down starting with the word "instance". An instance is nothing more than a virtual server in the cloud, running Windows or Linux just like any other server in your dusty server closet. While the instance is running, you accumulate usage charges--turn it off and the meter stops. The hourly rate you're charged varies based on the size and power of the underlying hardware and it can change from time-to-time and from region-to-region. This is called the "On Demand" rate. And while On Demand rights compare quite favorably to buying and operating your own servers, you can get a much better deal if you are willing to commit to using that virtual server (i.e to "Reserve" it) for 1 year or more.
Three Questions You Need To Answer:
1. What are the pros and cons of reserved instances?
There are many pros that come from reining in your AWS costs. You'll be able to scale effectively without breaking the bank, and you'll find ways to unlock the potential of cloud-native apps like artificial intelligence and machine learning and you'll do it all faster than ever before.
On the other hand, if you didn't really need that server for a significant part of the year then you would have been better off to pay On Demand rates and have just turned it off when you weren't using it. There is a break-even point that's easy to calculate, you have to be reasonably confident you'll reach it.
2. Which types of clients / users typically see success using reserved instances?
Success with reserved instances is widespread. Companies engaged in a full cloud migration benefit from reserved instances. Managers who are battling to understand and predict their costs also find it enormously useful. Organizations that are growing fast and need maximum flexibility see success. For example, online payment processors Stripe explain how they use AWS Reserved Instances 'to predictably forecast our cloud spend given a dynamic fleet with rapidly changing compute requirements.'
3. What does “success” look like in using reserved instances?
Using reserved instances brings a sense of control and predictability to your cloud services. You'll find your AWS bill is more predictable, you're not paying for unused resources, and there are no nasty surprises to trip up your AWS cloud migration.
There's a lot more to learn about reserved instances and AWS cost optimization but hopefully this introduction will give you the confidence to take the next step. Unfortunately, we can't promise that your home life will be any more manageable, but at least you'll be secure in the knowledge that your AWS cloud costs are under control.
For more information about Actionable Implementations, read our full guide on AWS Cost Optimization.