Ep. 127 | Savings Plans Overview & Exam Prep | Cost Mgmt | SAA-C03 | AWS Solutions Architect Associate
Chris 0:00
All right, cloud engineers, buckle up. Today. We're diving deep into AWS Savings Plans.
Kelly 0:05
Yeah, you know, this is one of those services that can really make a difference in your cloud costs,
Chris 0:10
absolutely, and especially for anyone listening who's working on, you know, big cloud projects, absolutely.
Kelly 0:14
So
Chris 0:15
let's, let's get right into it. What exactly are savings plans? I mean, I know we used to talk about reserved instances, right? But this is, this is something different, right?
Kelly 0:24
Yeah, so savings plans, think of it like a commitment based discount. Okay? You're basically saying, hey, AWS, I'm gonna use this much compute power over a certain period of time, and in exchange, they give you a discount on that usage. So instead
Chris 0:41
of reserving a specific instance, I'm basically pre paying for a chunk of compute power exactly, and then I can use that power however I want, like across different services. That's exactly
Kelly 0:51
it. And this is where it gets really interesting. We're talking EC, two Fargate, even Lambda.
Chris 0:57
Wow, that's a lot of coverage. So I'm not locked into just one service or instance type.
Kelly 1:01
Nope, a lot more flexible than reserved instances in that way. That makes sense.
Chris 1:06
So I'm getting discount. I've got flexibility. It sounds pretty good, but what kind of savings are we talking about here? Well, you
Kelly 1:11
can potentially save up to 72% off the on demand prices.
Chris 1:15
72% that's that's huge. It is. Can
Kelly 1:17
make a big difference, especially if you're running large scale applications. Yeah, no
Chris 1:21
kidding. All right, so we've got flexibility big savings. There's got to be a catch, right?
Kelly 1:28
Well, there are definitely some things to consider, like, there are different types of savings plans, and each one comes with its own, you know, set of pros and cons. Okay,
Chris 1:38
I'm sensing a decision point here. Definitely walk me through the different options. Maybe give me some examples of when I choose one over the other. Sure. So
Kelly 1:45
let's start with Compute savings plans. These are the most flexible. You can change instance, families, operating systems, even regions, and still keep your discount. So
Chris 1:53
if my workloads are constantly changing, maybe I'm using a mix of different services. This would be a good choice. Exactly.
Kelly 1:59
It's ideal for dynamic workloads, where your needs might shift over time. Makes
Chris 2:04
sense. Now, what about the other option? What was it called, again?
Kelly 2:07
EC, two instance savings plans, right?
Chris 2:09
EC, two instant savings plans. What's the trade off there? Well,
Kelly 2:13
with instant savings plans, you get the biggest potential discounts. Yeah, okay. I like the sound of that, but they're tied to a specific instance family and region. Ah, so less flexibility, right? So if you've got a stable workload, you know exactly what instance type you need, and you're going to be using it long term, then instant savings plans can really pay
Chris 2:33
off. So it's really a matter of matching the savings plan type to my specific workload needs
Kelly 2:38
exactly. Analyze your workloads, figure out your tolerance for commitment, and then you can make the most strategic choice.
Chris 2:44
All right, this is starting to make a lot of sense. Now I'm curious about the bigger picture. How does savings plans work with, you know, other cost optimization tools, like, I'm thinking reserved instances.
Kelly 2:55
Yeah, good question. So savings plans are actually designed to work alongside other tools. They're not, you know, mutually exclusive. So I
Chris 3:03
could potentially use reserved instances for some workloads and savings plans for others,
Kelly 3:07
absolutely like, say, you have mission critical apps that need a guaranteed level of performance. Reserved instances could make sense there and then you could use savings plans to cover your more dynamic workloads, maybe those running on Fargate or Lambda. Okay, so
Chris 3:22
it's all about creating this kind of multi layered cost optimization strategy precisely.
Kelly 3:27
And this is where it gets really interesting, you know, because it's not just about memorizing facts. You have to analyze your needs, understand the tools, and then get creative in how you combine them. That's
Chris 3:37
what makes this feel so exciting, right? Exactly. All right. This is a great foundation. Now I want to get into some exam prep scenarios really test our understanding of savings plans.
Kelly 3:47
Sounds good. Let's see what we can come up with. All right, so let's say you're working with a company and they're moving a large application to AWS. This app, it's going to be running on EC2 instances, and they're planning to run it for like, at least three years. Their workload, it's pretty stable, not a lot of fluctuations. What kind of savings plan would you recommend for them?
Chris 4:08
Okay, let me think so, long term commitment, stable workload. Sounds like EC2 instance savings plans would be a good fit
Kelly 4:15
exactly. They know what instance type they need, and they're in it for the long haul. Instant savings plan lets them lock in that sweet discount, perfect,
Chris 4:24
maximum savings, and they don't really need the flexibility of compute savings plans anyway. All right, give me another one. Okay,
Kelly 4:31
this time, we have a startup, and they're launching a brand new e commerce platform on AWS, but they're expecting some big traffic spikes, you know, like during the holidays or special promotions, right? And they're also planning to, like experiment with different instance types as their platform grows. What do you think? Hmm,
Chris 4:49
that's tricky. Sounds like they need something more flexible. So I'm thinking compute savings plans would be the better choice. Here. You got
Kelly 4:56
it. They need that flexibility to handle those traffic spikes. And to try out different configurations with Compute savings plans, they can scale up and down, move between services, all while still getting a discount. Makes
Chris 5:08
sense. Yeah. Now let's say we've got a company running a really important application, mission critical stuff, and they're already using reserved instances for that, but they also have some other workloads that aren't as critical and those workloads, they kind of fluctuate a bit. What's the best strategy there?
Kelly 5:25
Ah, a multi layered approach. They could definitely use savings plans to cover those variable workloads on top of their reserved instances for those mission critical apps. It's
Chris 5:34
like they're getting the best of both worlds, right,
Kelly 5:36
savings and flexibility. Okay, I'm
Chris 5:38
feeling more confident about these different scenarios. But what about the exam itself? Like, what other types of questions might they ask? Well,
Kelly 5:45
they could get into the financials, like asking you to calculate potential savings or compare different savings plans. So we'll need to know some formulas and stuff like that, yep, things like break even point, effective hourly rate, you know, being able to analyze those numbers, right, right. Okay. What else they might ask about the different commitment links one year
Chris 6:05
versus three years
Kelly 6:06
exactly. You gotta be able to explain the pros and cons of each option. Got
Chris 6:10
it so understanding the technical stuff is important, but we also need to know how to apply that knowledge to make smart financial decisions. Absolutely.
Kelly 6:19
And don't forget about how savings plans fit into the overall AWS ecosystem. They might throw in questions about AWS Budgets or Cost Explorer. Yeah,
Chris 6:29
all those tools could be used together for a more holistic cost management approach. Exactly.
Kelly 6:33
It's about seeing the bigger picture. This is great stuff.
Chris 6:37
I'm feeling much more prepared now. Glad to hear it.
Kelly 6:39
But hold on, we're not quite done yet. There are still a few more things to cover. All right. So
Chris 6:43
we've talked about the different types of savings plans, how to choose the right one, even touched on some financial calculations. What else do we need to know?
Kelly 6:51
Well, let's talk about how savings plans actually work, like how the discounts are applied to your bill. Okay, yeah, I've been kind of wondering about that. So savings plans, they work on an hourly basis. Every hour, AWS looks at how much compute power you're using, you know, across all the eligible
Chris 7:06
services, right? EC, two, Fargate, Lambda, yep, all of those.
Kelly 7:10
And then they calculate the cost of that usage, but at the discounted savings plan rate up to the level you committed to. So
Chris 7:19
let me see if I've got this straight. Let's say I have a savings plan for like, $100 of compute per hour. Yeah, if I use $80 worth in an hour, I get that full discount. But if I go over, let's say $120
Kelly 7:31
then, then you get the $100 discounted, and that extra $20 that's billed at the on demand rate, makes
Chris 7:36
sense. It's like I've got this pool of discounts that refills every hour Exactly. And
Kelly 7:40
this is a key difference from reserved instances. With reserved instances, you're paying for that instance whether you're using it or not, but with savings plans, you're only paying for what you use up to your commitment. So
Chris 7:53
savings plans can be more cost effective if my workloads are kind of unpredictable, right?
Kelly 7:58
If you've got variable usage patterns, savings plans give you that flexibility without sacrificing savings
Chris 8:04
got it. Now you mentioned earlier that savings plans can be used with other cost optimization tools. Can you give me some examples of how that might work?
Kelly 8:13
Sure you could combine savings plans with reserved instances, for example.
Chris 8:17
So like reserved instances for those steady always on workloads and savings plans for everything else, exactly,
Kelly 8:22
and you could even throw in some Spot Instances if you have workloads that can tolerate interruptions. Wow,
Chris 8:29
so many options. It's like building a custom cost optimization strategy. That's
Kelly 8:33
the idea. It's about finding the right combination of tools and techniques that works best for your specific needs. This
Chris 8:39
is really helpful. I feel like I'm starting to see the bigger picture, not just about savings plans, but about the whole cost management landscape in AWS. That's what
Kelly 8:47
it's all about. And remember this stuff, it's constantly changing. So always keep learning, keep exploring, and you'll be well on your way to cloud cost mastery.
Chris 8:56
Well said. And on that note, we're going to wrap up our deep dive into AWS Savings Plans. Thanks for tuning in, everyone.
