AWS provides a large pool of resources designed to be flexible in a highly scalable environment. But with all of these resources available upfront, it is easy to get excited and go overboard designing your infrastructure. What AWS doesn’t provide is a way to keep costs in check, optimizing the precision of your AWS environment is essential to increasing the value.
At Mission, we’ve worked with our customers to transform their traditional on-premise IT setups into efficient AWS cloud based infrastructures. With many tools available like nightly server AMIs, snapshots directly to S3 and no physical limitations, there are many benefits for IT in the cloud.
Acquiring new physical hardware on premise can quickly be expensive with high upfront costs and continually maintenance. Plus any systems you install will be limited by hard drive space, CPU, and RAM installed making upgrades a challenge. With instances in AWS, you can design each instance to exactly the resources you need with easy ability to upgrade at a later time – but you pay for exactly what you set up, not what you use.
For example, you can run a Domain Controller on an m3.2XL that is only utilizing 5% of m3.2XL, but you’re still paying top dollar for the 2XL. So when working with AWS, a best guiding principle is: start small and build up.
When launching and building instances, start with the smallest instance size possible and work up. You don’t need to know the exact specifics when launching and can easily upgrade at a later time.
If you need to set up a server for your in-house application, you can easily set up a Windows server but until you know how powerful it needs to be, you can launch as a t2.micro and bump it up as necessary. AWS’s design makes it exceptionally easy to switch instance types, so use it to your advantage!
Here’s the price difference between running a t2.small versus an m3.large (Windows):
t2.small: $0.036 hourly or $0.864 a day
$315.26 a year
m3.large: $0.266 hourly or $5.42 a day
$1978.30 a year
If you aren’t using your m3.large to it’s maximum potential, you could be throwing away $1500 a year in unused performance per instance!
Want to know more about AWS Cost Saving Commandments? Download our Cloud optimization Datasheet!