E01: WELCOME TO THE CLOUD: The Cloud Cost Optimization Podcast Kick-Off

Hey, what's up everybody? This is Nick from 10 nasty cloud.com with my co-host Jason, and you are listening to the Cloud cost optimization podcast. More people than ever before, building cool stuff in the cloud and spending a lot of money doing it. On this show, we discuss cloud optimization trends. Deep dive into the technology, talk to experts living in the cloud native hyperscale world to learn from their experiences and understand their strategies.

Welcome to episode one. In this episode, I'll be interviewing my co-host to learn a little of more about Jason's background and frame up the three pillars of a solid cloud cost optimization program. Jason, it's our first show. Uh, why don't you tell us a little bit about your background in, uh, technology and Cloud comput.

Yeah, I've been in cloud for pretty much my entire career. Now going on about 20 years, that's how old, uh, we are getting in this arena. So we were like in [00:01:00] its infancy, in cloud. Um, you know, so I've seen a lot of changes in the industry. Love, have, loved aws, Azure, and those types of public clouds since, since they've come out.

And, uh, really loved the benefits and all the flexibility they give you. Um, but we've seen over, and I've seen over and over again in my, um, in my past, How much this stuff can cost and how, uh, how badly the security can get out of control if you don't have the right tools and processes in place. So, um, that's a little bit about me.

Yeah. It, uh, I remember the early days of setting up my first, you know, ESX hosts and, what was it? Gsx was gsx before that, those hosts, and I mean, we had the same problems, like, like sprawl and, and cost and buying hardware. It was, it was all the same sorts of problems. You remember the first sprawl, I mean, there was sprawl before that, but when like virtualization came out, everybody just spun a virtual machine up for every application.

It was like the first. Just spin up a new virtual machine for this. It's like, and then, then all of a sudden you [00:02:00] had, you know, hundreds of virtual machines. Yeah. It's the, it's, they're the same problems just at, in, in, in, at a different scale that we've seen for every junk server we could find. Every single junk server we could find.

We threw, uh, uh, you know, cloud hosting software on, uh, that was, that was prior to the AWS and Azure days. But, uh, that, you know, Right. That sort of changed everything. So it did. Why, why focus on cloud cost optimization? What gifts? Well, I mean, you know, listen, for, for the last forever, for as long as I can remember, every year, every report that comes.

It's estimated that 30 plus percent of cloud spend is waste. And when you think about the revenue of, just say aws, uh, in particular 50 billion before the pandemic hit, 30% of that's a pretty big number. Um, I think they're gonna be close to 80 billion in annual revenue after the pandemic. [00:03:00] You know, we just saw, you know, people needed to get stuff up pretty quickly to support remote work and, you know, so we saw a huge rush into the public cloud during the pandemic.

So estimated to be 80 billion now and. The percentage of wasted spend is not estimated to go down. In fact, it goes up almost every year. And we think companies should be paying for what they're using and not , you know, just lining, lining the pockets of, of, uh, billionaires. Yeah. We like to, the, our pithy statement is, you know, don't get ripped off by your cloud provider.

Right. You know, they're not telling you where the waste is. The, uh, they tell you waste has time. They tell you specifically what they want you to. That's it. What they want you to save. That's what they tell. That's right. That's right. Waste has this, uh, time function to it where, you know, it's, it's really, it's an, it's an over time, you know, sort of expansion and there's, there's these thresholds that we see, you know, when an organization gets to about [00:04:00] one year in their cloud journey, two years into their cloud journey mm-hmm.

uh, you can start to find really significant. Uh, kind of abandonment of, of infrastructure and it, it's from a number of different places. Uh, one is just rate a change. I mean, just, just dozens, sometimes hundreds of changes a day in these organizations results in things that just get left behind. It happens, uh, or things that are configured incorrectly.

Uh, that, that's a fairly common issue. Uh, you know, two is that, um, legacy, uh, infrastructure. As you know, cloud deployment patterns evolve because they almost, uh, certainly start at a very simple, uh, uh, deployment strategy. Uh, when you're going through digital transformation, when you're going through your cloud migration and then you, you start to discover the power of cloud and you adapt and evolve your.

Your cloud deployment, cloud infrastructure, a lot of legacy infrastructure can get left behind. Sometimes it's the safeguard, like, you know, well, let's just leave it there just in case and we'll, we'll delete it in 30 days, I [00:05:00] promise. Cross my fingers. But it doesn't, you know, that cleanup oftentimes just never materializes, uh, or that infrastructure has forgotten and it lays around so, so waste sort of gathers over time this time.

um, you know, out outside of waste. Uh, I think that there is, you know, significant opportunity for savings in, in really, you know, managing committed use discounts. And, and I, I, I think this is, you know, the, I guess the right word for it is confusing. It's oftentimes confusing to the end user and we, you know, we see, uh, in, uh, our deployments.

Oftentimes, you know, only around 10 to 15% of of consumers are are actually making some sophisticated use of, of committed use discounts. They, they may be, you know, there's, there's a broader percentage that are, you know, potentially picking a committed use discount. But, uh, they're, they're using a very simple strategy of just throw one, you know, throw.

You know, savings plan at it. Uh, pick the simplest one and go, uh, you know, they try not to get [00:06:00] caught up in the analysis paralysis, which, which a lot of people do, but really only 10 to 15% of of consumers are actually, you know, having committed use discount strategy. Can you talk a little bit about that difficulty and why there's confusion?

Yeah. I mean, it, it has to do with a lot of those, uh, committed, um, Uh, committed discount, uh, vehicles that the providers have are based on information that hasn. Necessarily been available to say the fin ops team or the cloud center of excellence or cloud cost optimization team? Right. So as an example, reserved instances are, uh, you know, specifically for persistent workloads, things that use a lot of compute all the time.

And so you, you tend to need contextual information to select the best commitment vehicles. Uh, vehicles, by the way, it's, it's, it's typically more than one in all of them that you should be utilizing. [00:07:00] And so what oftentimes happens is that companies choose to deploy the easiest one, right? The one that covers the most broad category.

It's the easiest. It makes you feel good. You save 10 or 15%. Heck, it makes the provider feel great too because you're only saving 10, 10 or 15%, and they know if you really unlocked the true power of these vehicles, you could be saving 40%, uh, up to 40%, maybe even more in some cases. It just depends on what your workload is like, and, and it, so, so you need to kind of have that contextual information married so that you can underst.

usage, workload type, how these things are being utilized. And you also need to go through and do some optimization first because you don't wanna buy discount vehicles on overutilized things or underutilized things, right? It's not gonna be, So you need to have thefor, you need to marry the technical, the usage information with the cost information to make the best decision on all of the commitment vehicles to use.[00:08:00] 

And it, it is. It's really difficult to do with the tools you're given. Like I said, AWS gives you all the tools to save this money, but they show you what they want you to save. It just is what it is. We, we would expect it. And so, um, that's the reason why. . Well, and you have to be even, even more careful with, with Azure, it's not as sophisticated.

There's not as many tools to use from a committed, used discount perspective. But there's also not a mature marketplace, uh, by which you can, you know, buy and sell and trade when you, when you, uh, need to, need to change 'em. So, uh, you know, there it takes a little more forecasting and a little more for knowledge, a little more tuning up front to select the right plan and make sure that you're maximizing your savings from an Azure perspective as.

Um, so you and I have touched on, you know, the, uh, uh, in the past, I think it's important to cover here the, you know, really what, what are, what's the foundation of a good cloud cost optimization program? You know, what are those three pillars? And, you know, it, it [00:09:00] starts with, you know, really understanding everything you have.

What are all the, the puts and takes? What are the inputs to the system? What creates change? So having that kind of, you know, big picture, what. All is there and it, it's kind of surprising, I know you'll touch on this, but it's kind of surprising how often folks don't have that big picture in their organization.

They, they really are kind of blind to, to parts of it or what's going on in parts of it. Two is, you know, the accountability side of it is who owns what pieces and is therefore responsible for it. So, you know, you kind of. Got this cost allocation, charge back, billing bill, back showback, uh, uh, component to it, or from a partner perspective accurately invoicing customers.

That's, that's, we've seen that be an issue over and over and over in our, in our conversations and in our deployments. Um, in, in the third piece of this is really just making use of the right. Tool sets in the right strategies around, uh, managing, you know, discounts and, uh, cleaning up waste, abandonment, idle, you know, rightsizing resources and so on.

Uh, you know, uh, we've seen folks go from having a baseline of a few percentage points or even 10 or 15% savings in their environments to actually being able. To manage 30, 40 plus percent, uh, savings ongoing because they have a sophisticated strategy around how to actually, you know, get those discounts and manage that environment.

Um, tell us a little bit about, sort of, you know, your thoughts on, you know, what you have to tackle first. What are some of the lessons learned? Uh, in, in, you know, seeing this in the wild all starts with visibility. Um, you know, if, if it's really difficult to implement a pretty, a good commitment management strategy, if you don't know what you have and we're not, and you're not working with a base set of infrastructure that you know you need, um, and, and honestly, uh, it is something that we see often is that, you know, Show me a list or show me where I can see everything that you've deployed [00:11:00] across all your organizations, accounts, regions, subscriptions, tenants, whatever it may be.

And, uh, that's not easy to do, and that's the basis of what you need to make sure that one you have. Just the things that you need out there, right? Because we don't want to be paying for infrastructure we're not using so that you can move on to the next steps, like optimizing for the right types of storage, compute cetera, and then onto commitment management because you don't want to be.

Committing to, to, to compute usage is an example that you don't need. You could be measuring a baseline of things. Your baseline could be 20% higher than what you actually need because you have abandoned resources or you have underutilized resources. So it all starts with visibility. It's very simple to do.

You can do it in five minutes from our website and have a manifest list of everything that you have in your environment. That's the first thing to do. I think, [00:12:00] um, uh, you know, that's, that's a really good point is, is there is kind of a, a strategy one to, you know, three or, or one to maybe even five or 10, you know, steps in this process of, of actually getting it right and the more closely aligned to that, uh, strategy in the fas, you move through it, you actually.

Get to potential savings much, much faster. So, uh, you know, I think organizations that kind of get stuck, you know, just, just getting the visibility piece right, you know, and there's, there's all sorts of tools out there to use to, to get visibility, but it's really important to nail that right off the bat, understand what you have so you know what to clean up before you go, start committing to, to uh, uh, you know, commit committed discount program.

Um, well, I think, I think that wraps our show for today. Uh, so thank you so much, Jason. Uh, uh, it is fun, uh, being able to have this conversation actually, uh, uh, on recording. These conversations happen all the time between us, but, uh, we should just record all of our conversations, Nick. We should all of our wanna listen to that.[00:13:00] 

It'll be a different podcast. What would be the card cost optimization podcast? be very different podcast. A whole lot of debates and arguments, uh, which is what found to do. Yeah. So, uh, anyway, that's it for today. Uh, tune in. Uh, To, uh, the cloud cost optimization podcast. We'll be dropping content twice a week.

Uh, we look forward to your feedback, uh, and uh, to have some of you all on the, on the show as well. So thanks. Keep tuning in. Uh, go up to Tennessee cloud.com for your free assessment. And, uh, we, we would love to have the chance to show you how to optimize your environment.