AWS Free Tier

AWS has a very famous free tier, broken down into three categories:

  • Free Trials
  • 12 Months Free
  • Always Free

Free Trials

These services are free for a certain period of time after you first use them.

So towards the middle of the course, we'll enable secrets manager, which will be free for the first 30 days we use it. After that, we will have to pay the normal rate for secrets.

From the AWS docs:

Free Trial

30-DAY TRIAL PERIOD

You can try AWS Secrets Manager at no additional charge with a 30-day free trial. The free trial enables you to rotate, manage, and retrieve secrets over the 30-day period.

Your free trial starts when you store your first secret.

Pricing

PER SECRET PER MONTH:

$0.40 per secret per month. For secrets that are stored for less than a month, the price is prorated (based on the number of hours.)

PER 10,000 API CALLS:

$0.05 per 10,000 API calls.

Always Free

This usually means that the first amount of usage or storage within our account is free, no matter what.

So if we have up to 25GB of storage in DynamoDB, or 1 million lambda requests per month, we're not going to have to pay for those. And if we exeed that, then we pay the standard pay-as-you-go rates.

For lambda, that would be $0.20 per 1M requests after the first 1M.

12 Months Free

This is only available to new AWS customers, and is available for 12 months after you sign up.

After 12 month or if you exceed the free tiers, you pay the standard pay-as-you-go service rates.

EC2 Compute Time

For example, we get 750 hours of EC2 compute time for free each month, for the first 12 months.

Since there's just under 750 hours in any month, this means that we could leave an EC2 isntance running 24/7 for a year and it would cost us $0.

But the time is per month, so we could have two instance running for half a month, then terminate them for the last half of the month. Then the next month we could have 4 instance running for a week, then terminate them, and we still wouldn't have to pay for EC2 compute time.

And this is the same for RDS and other services

Don't Pay Too Much

Most of my students tell me they end up paying between $5 and $10 on AWS infrastructure in order to complete this course. I've also had a few students with a $300 bill at the end of the month.

All you have to do to avoid a large bill, is terminate your infrastructure when you're not using it.

After each deployment, I will explicitly tell you what you should terminate, stop, or leave running. If you follow these instructions, you won't end up with a surprise bill at the end.

Show timestamps
00:00
AWS has a very generous free tier that's broken down into three categories. We have free trials,
00:06
twelve months free, and always free. We can scroll down here and see which services
00:11
fall into each category. So if we look at free trials, I'm actually going to search for something
00:15
that we're going to be using, which is Secrets Manager. You can see that things in this category are
00:20
free from the moment you start using them. In this case, we would get thirty days free. So
00:25
at some point in the course, we're going to set up Secrets Manager, and for the first thirty days of
00:29
using that service, it will be completely free. Then after those thirty days, if you still want to use
00:34
it, if we scroll down to pricing, we can see what the actual cost ends up being. So forty cents
00:39
a month per secret, with some cost for API calls. Keep in mind that some services will be free from
00:45
the moment you use them, but just for a limited amount of time. Then there's the always free category, and
00:50
these are awesome because no matter what you've already done with your account, you will always
00:55
get some amount of storage or compute for free with these services.
00:59
So if we use DynamoDB, we get the first twenty-five gigabytes of storage free. If
01:04
we use AWS Lambda, we get the first one million requests completely free within a single account.
01:10
And then if we go over that first million, if I go over to pricing, I think with Lambda
01:16
it's going to be really cheap—yeah, it's only twenty cents per one million after that. So we're saving
01:20
cents, but that's still nice. And then the final category here, which is probably the most famous, is
01:24
the twelve-month free tier. So from the moment you set up a new AWS account, you get twelve months
01:30
free for certain things—a certain amount of usage of certain services.
01:34
And I do suggest for this course setting up a brand new account for this reason. It also sets
01:40
up your account in a way that works really well with the course. So if you do that, and you
01:44
have a brand new account now, you'll get twelve months free of many of these services. Now, EC2
01:49
is going to be one of the most common that we talk about and use throughout this course.
01:53
You get 750 hours per month of free EC2 usage. So if we look at that
01:58
quickly—if there are 24 hours in a day, and let's say there are 31 days in a
02:02
month—this gives us just over one month of free usage of an EC2 instance. So we could set up
02:08
one EC2 instance—that's one virtual machine—and leave it running for a whole year, the first
02:13
year, without having to pay any money. After that, we'd have to pay the normal pay-as-you-go
02:17
rates. But this also means we could have two EC2 instances running for half a month,
02:22
terminate them for the second half, and do the same every month, or have four EC2
02:26
instances running for about a week, then terminate them all—and it resets each month. So we can
02:30
kind of mix and match that. For some deployments where we need more than one EC2
02:35
instance running, that's fine. It's still within the free tier as long as we try not to use up
02:38
the 750 hours that we get for free each month. Then we get another 750
02:43
hours of RDS, we get a certain amount of S3 storage, and as much as possible,
02:47
we're going to stay within the free tier. So this course will not cost you much money at all. There
02:52
are certain services that you just have to pay for, like Route 53—there's no free tier
02:56
even though the EC2 instances get 750 hours for free of compute.
03:01
The IPv4 addresses now cost money, so there'll be a small charge every once in a while, but
03:05
students who've been through this course say they usually spend around five to ten
03:10
dollars to complete the entire course. So it's not a lot of money, but I've also had students
03:15
complain that they've spent up to three hundred dollars in a single month—and that's
03:19
generally because they spun up a whole bunch of EC2 instances or RDS instances and
03:24
then forgot to terminate them. But throughout the course, I do note when you're supposed to
03:29
terminate some infrastructure, so if I scroll to something like the RDS section, there's a note right
03:34
here saying make sure you fully delete the RDS instance. So as long as you're following along and
03:39
using some common sense, terminating your infrastructure when you're done with it, then you
03:43
shouldn't spend much money. And if you find yourself coming to a part of the course like setting up
03:48
an RDS instance and you haven't yet completed that section, and you know you're not going to be back for
03:52
a while, just terminate the infrastructure. It's fine—it's the cloud, you can re-set up that
03:55
infrastructure later on. It's good practice and you're not going to spend any extra money. So really,
04:00
be careful if you're setting up something that can cost money if you go over the free tier usage—
04:04
just make sure you're terminating that when you're done using it.