Podcasts

How Druva Reduces Your TCO

W. Curtis Preston, Chief Technology Evangelist

Druva often makes the claim that it reduces customers’ TCO by up to 50%, but just how does it do that? Two men who understand backup and resiliency costs more than most answer that question. W. Curtis Preston (AKA Mr. Backup) and Stephen Manley (CTO), talk about the hard and soft costs that Druva reduces. The discussion focuses mainly on hard costs, as many customers dismiss soft costs. But Curtis and Stephen discuss soft costs as well, as they feel they are part of the TCO story as well.

[00:00:00] W. Curtis Preston: This week on No Hardware Required. We’re talking about saving money with your backups. With me as always is Stephen Manley, our CTO. Thanks for joining.

Hi and welcome to Druva’s No Hardware Required podcast.

I’m me host W. Curtis Preston. And with me is my TCO expert. Stephen Manley. How’s it going, Stephen?

[00:00:23] Stephen Manley: It’s it’s going great. You know, I, I, one of my personal heroes is Benjamin Franklin and, you know, he famous for saying. A penny saved is a penny earned, which with interest rates up until recently, it was really only a penny. But now Hey, interest rates, it might be a penny and a little bit of money.

[00:00:41] W. Curtis Preston: There, there might be something there as long as it’s not in stocks, but that’s a whole other thing. Um, not, not the time to be in

[00:00:48] Stephen Manley: it’s a half penny. It’s a half. Or if it’s in Bitcoin, it’s a 10th of a penny. Anyway.

[00:00:57] W. Curtis Preston: I thought we’d take a little bit of time to talk about, what it’s like to save money in backups. And when we talk a lot about how we save customers money when they move from a traditional backup and recovery system to us. And so I thought that would be a good topic for a podcast.

And I would, I’m just gonna throw out to you. What do you think is the first area. When you’re talking to a potential customer and they say, well, well, how do you save money? What would be the first thing that you would, what you would think about.

[00:01:29] Stephen Manley: Oh, the, the first thing everyone brings up is. There there’s two parts to the discussion. The first is, uh, this had better be hard costs. I wanna see where you’re going to save me actual expenditures. We can talk about soft cost, management costs, afterwards, but you better be coming in with, with, you know, real sort of savings I can see in terms of what I’m putting on my, my budget. Uh, and then the second one is, and I wanna see that savings in a year. You know, don’t gimme any of this in three years, I’ll save money, kind of math. I want the math that says I save money this year because let’s face it. The economy is bad this year.

That’s when I need to save money. You know, two, three years from now, hopefully things have turned around and that’s all well and good. I need to save money now.

[00:02:17] W. Curtis Preston: Yeah, that makes a lot of sense. And I know. When we talk about, uh, I, I think. It was maybe two podcasts ago, maybe one podcast. I can’t, I can’t keep track of all the podcasts even, but I know that we talked about challenges with a traditional backup and recovery system. We talk a lot about security and we talk a lot about over provisioning.

Right that I, if you’re buying a backup system today, you, it it’s a capital expenditure. So you have to buy what you need for the next three years. You also have to over provision because a backup system is a really weird system that gets really busy for only a few hours of the day. And then, you know, and then, uh, and maybe even more busy at certain days of the week and then busier, also, if you go to do a large restore and you have to over provision for growth, over provision for that up and down nature, uh, over provision for just the sake of not having to go back to your boss? Um, I, I, I remember golly, gee, Stephen, a really long time ago, having to go back to my boss. We were out of backup tapes, right?

We had not properly provisioned the amount of tapes that we needed. And I remember having to go back to my boss and buy. It was something like $10,000 worth of backup tapes. You know, 30 years ago was an awful lot of money and now it’s, you know, lunch. I remember asking, you know, it’s that horrible moment when you have to go and ask for something, I thought you told me, you told me this was good for three years.

And so I, I just think it’s human nature to just over provision, to, to deal with all of those things. And I know you’ve worked a lot with customers before this company. That you were probably very happy for them to over provision. So, you’ve seen this first hand.

[00:04:21] Stephen Manley: Yeah. I mean, there, there were a lot of systems that I’ve worked on in the past from, from multiple companies, multiple appliance companies, I’ve worked at where you would look at the phone home supports and you should be like 80% of these systems never break like 10% utilization, at least from a, from a CPU, memory IO kind of performance, uh, standpoint, and a number of the systems on the backup side never broke 30, 40, 50% utilization.

Because like, like you said, there is nothing worse, uh, that than going to your management, who has already sort of done their, their capacity planning for year, they’ve done their budgeting and going and saying. I was wrong and I need a bunch more money because that is so much worse than asking for a bunch of money at the beginning.

And, and so, so yeah, we see lots and lots of people who are just vastly over provisioned and, and, and to your point, the, the other interesting thing that we see though, is, is that tension between most people size it for their backups. So when it comes time to do that, heavyweight restore, especially a big ransomware recovery, they’re not scoped for that.

They actually they’re, they’re under provisioned for, for, for that doomsday scenario. So in a lot of ways, it’s the worst of our worlds where you go to your boss, you say, well, we were vastly underutilized, but now that we’re in this restore scenario, we’re actually overutilized. And so, yeah, I, I probably should just get another job at this point.

[00:05:45] W. Curtis Preston: Yeah. And, and so I, so I think that just a lot of people just, you know, they, they, they, they don’t want to have that meeting. And so the smart ones really do over provision, because it’s really the only way to have a decent backup system that is good at both backups and restores. I think very few people do just like what you said.

Very, very few people do actually over, over provision enough to handle a really large restore, but the ones that do waste even more money. Right. Because it it’s like having a, a Ferrari in your driveway. Uh, and working from home, I think it’s you only use your Ferrari to drive to Walmart.

[00:06:28] Stephen Manley: Exactly. Well, and, and, and the other part, I think that, that, again, happens in any corporation is, and, and we see this with schedules. We see this with, with, with, with budgets, everything is, you know, Curtis over provisions by 10%. Curtis’s boss adds another 10%, Curtis’s boss’s boss adds another 10%. And so even no matter how much you’ve over provisioned, it’s gonna be so much worse by the time it actually gets to, you know, to, to, to sort of the person issuing the PO.

[00:06:56] W. Curtis Preston: So how, how is it different with us?

[00:06:59] Stephen Manley: Yeah. That’s, you know, it’s interesting, you know, one of the common conversations I’ve had with virtually every customer I meet is they start with, you know, I look at a say maybe per-terabyte basis or some basis. And to say the cloud’s not cheaper than, than my on-premises infrastructure. It’s it’s about the same.

And you look and you say that may actually be true. The difference is when you buy a Druva, when you’re, when you’re actually getting an on-demand, you know, sort of pay for what you use kind of system. You’re only paying for what you use. So that 10% utilization means you’ve just saved 10 X, uh, compared to where you were.

So it’s not that, you know, necessarily it’s even raw materials are cheaper. It’s just that you’re not paying for unused, unused, uh, you know, cycles.

[00:07:49] W. Curtis Preston: Yeah, that’s a, that’s a really good point because we talked about over provisioning and that’s over provisioning based on sort of, I, I think I’m gonna have this, so I’m gonna build something like this and then I’m gonna build it bigger because of that. But in reality, a really significant portion of the time.

A really small portion of the actual, because you, because of the way you, um, you provision a file system. When you provision a file system, you have to provision the whole thing. So for a significant portion of the time, A a like vast majority of that, whatever you bought is not utilized. Whereas in the cloud world, the same would be true, except you didn’t do that provisioning.

Right. You don’t do provisioning with Druva. You simply. You know, we, we work out with you, you do pay in advance, right? Of, of what you, what we think you’re going to be backing up. And then we true it up every month. Right. We’re basically crediting against your account. And then the, the way I describe it is at the end of a year, one of two things happened either you bought too much or you didn’t buy enough.

Right. So if you bought too much, we just roll it over into the next year. If you didn’t buy enough, you just renew early. But the difference is you don’t ever pay for something that you don’t use, which is never true in the data center, right?

[00:09:16] Stephen Manley: Yeah, it’s, it’s one of the things to, to your point on that three year, I, I did, I did have a customer who came to me and said, you know, we, we actually, we nailed our provisioning this time. We, we got it exactly right. At the three year period, the system hit the 80% utilization, cuz generally with a dedupe system, you don’t wanna give about 80% cuz things go horribly wrong.

And I said, that’s great. So at the halfway point, you were at 40% utilization, you were still paying for the, at the one year point. Right? So, so even when you get it right, like you’re saying along the way you were wrong almost all the time until the very end when you were right.

[00:09:53] W. Curtis Preston: right. That, but that’s just the way, uh, you know, provisioning works, right? Um,

[00:09:59] Stephen Manley: It’s the way hardware is.

[00:10:00] W. Curtis Preston: it’s like a, it’s like a guy that’s gonna have a five, you know, he’s gonna have five kids, but he has to buy a minivan now because , and which means that the Van’s gonna go, you know, you’re gonna get minivan gas mileage for five years when it’s gonna be five years at, at a minimum before those five kids show up.

I think there’s another cost that we can talk about, which is things associated with ransomware. How, how can we save money there?

[00:10:28] Stephen Manley: Well, I tell you, I mean, one of the things I think that is different today than five years ago, and I think this gets to hard and soft costs is five years ago, you know? People looked at that 3, 2, 1 rule. And they said, okay, well first not everyone followed it, but even those who did said, look, I’m, I’m gonna get a copy on my disc.

And then if I really want that, that, you know, sort of separate media, maybe I’ll do a tape copy. Right. I, you know, I have some legacy tape infrastructure. I’ll do a tape out and, and that sort of ticked the box for them. But deep down inside, None of us really thought we were gonna use those tapes. Uh, and if we did, it was gonna be once in a blue moon and it was gonna be sort of, I need to pull something back for a legal case and, and, and it’s not gonna be sort of a, a high stress time, time constrained environment.

And then ransomware came along and everyone realized, well, ransomware, especially once it started targeting my backups, I actually do need another copy. Completely separate, ideally separate accounts, air gap networks, the whole deal. And we saw a bunch of people looking, saying, So I have to like buy another appliance or I have to set up my own cloud storage or either way it was yet another copy of the data.

And, and I don’t care where you are, how much dedupe you get, making another copy of the data is expensive.

[00:11:50] W. Curtis Preston: And yeah, it, it, depending on which service we’re talking about, right. You’re making another copy. You’re just paying for the storage. Maybe you’re also paying for another appliance, a virtual appliance in the cloud to be able to receive that copy. Maybe you’re paying for another physical appliance in another location, right?

Maybe you’re maybe you’re. You don’t wanna put it in the cloud. You wanna put it in a data center that you control. So you’re buying another physical appliance. Um, and there’s also a lot of costs associated with sort of. Preserving the data. So, so that people make that second copy. They’re also adding additional costs.

I just got an email this morning for a, another piece of software. I’m gonna call it a companion piece of software from one of our competitors. It’s, it’s actually not from the competitor, but it’s a piece of software designed to work with the competitor. That runs on Windows-based backup servers, and it was here.

Here’s another piece of software that you can pay for that will help protect your data from ransomware, because it’s not the, you know, that that’s not being provided by the backup software in question. Right? So it’s, it’s all of these extra things, extra copies, extra software, extra hardware. All of that adds up to your costs.

And so I think that, I think that goes pretty solidly towards those hard costs that you talked about. I do think it’s important to also talk about soft costs, because I spent, you know, the better part of 20 years helping people implement, re-implement, re re-implement their backup systems, because it was like, I, I remember, um, doing backup audits, right.

I would do a backup audit and, and, and it was like $10,000 for me to walk into the door and it was like, If I don’t find something wrong, it’s free. Right. Obviously I never gave it away because of the way backups have always worked. Backups are really complicated, right? Just the 5, 6, 7 different vendors that you end up messing around with that wanna point fingers at each other, the nonsensical ways that some of these devices, especially tape, uh, the way it behaves versus the way you think it behaves.

Um, and just, and the fact that you’re just sort of fundamentally fighting the laws of physics the entire time that it it’s, it’s not something that you can hand to, although we often do hand it to the new person, right. It’s how I got my job. Right back 30 years ago, somebody said, Hand the backups over to a, of a 35 billion company.

Let’s hand us over to this guy who got, who got access to his first root prompt three weeks ago. right. And so it’s really complicated. Uh, and, and so I think a lot of people there, we had a team. I mean, this is a few years ago, obviously, but we had a team of six people that were running the backup systems.

And I know that that’s the way it is at a lot of companies, maybe not six people, but they have a dedicated, really solid person. And this is maybe even more true today than before, because you need someone that also understands information security, that understands the concept of least privilege and, and, you know, um, things like multifactor authentication and all of those things that I didn’t know anything about back in the day.

[00:15:27] Stephen Manley: Right. To me, one of the, there was, uh, someone I used to work with who said backup is relatively unique in the it space. You know, maybe antivirus is a little bit similar in that it’s one of the few things in it that literally touches all of your different data sources. So if I’m a NAS administrator, more often than not, I only administer NetApp.

And so maybe I’ll administer NetApp box and an Isilon, if we have both of them, but that’s actually still relatively rare. If I’m a storage admin, maybe I’ll, I’ll administer my SAN, but I’m probably not managing the VMs. Right. And, and I’m certainly not running databases. And so. We expect the backup team to be able to connect to Oracle, SQL, Microsoft, Linux, uh, NAS systems, you know, now SaaS applications, 365, Salesforce cloud native, and, and, and yeah, to your point, we either go, well, geez, nobody really wants to do that.

So let’s give it to the new person. Good luck. Um, or you get a, a, you get some really senior people because the company realizes this is critical. You know, ransomware is important and disaster recovery is important and compliance is important either way. You’re either taking on a lot of risk or you’re taking one of your, your, your top paid key talent, and they’re focusing on this as opposed to some, maybe, you know, more business specific agenda to try to get you to cloud, to try to build new apps, to try to, to try to push your business forward. So, so yeah, those soft costs are real because this stuff is hard. If you try to do it by yourself. Right,

[00:17:04] W. Curtis Preston: Yeah, it’s really, it’s really easy to get it wrong, right. Evidenced by my, you know, near 20 year career and, in helping people put it back correct. Right. Uh, by the way, I can’t believe that we didn’t bring this up in the first half, but you, you mentioned DR right. From a cost perspective, there is no cheaper way to do DR than the way we do DR in that, um, you know, so you, there are a number of other ways to do it. Uh, well, other than not doing it right,

[00:17:37] Stephen Manley: Which is still, I think for a lot of companies, kinda their default of nothing bad’s gonna happen, you know, why nothing bad’s gonna happen. Cause I don’t want it to happen.

[00:17:46] W. Curtis Preston: exactly, you know, a box of tapes at Iron Mountain is not a DR plan. Assuming that that that’s not your DR plan, there’s no cheaper way to do it than to do it in the cloud. DR is the killer app for the cloud, because what you need is a ton of infrastructure and you need it instantly, but you don’t want to pay for it until you need it.

And that’s the way our DR system works. Right? You set it up in advance. There’s a little bit of pilot light cost to, to, to have it ready to go. But when you need to declare a disaster, you get instant, you know, use your phrase again, we can darken the sky on your behalf and we can restore all of your VMs simultaneously, because we have the power of the cloud at our disposal and then turn it all back on.

And that doesn’t cost. That costs you a fraction of the costs of doing this any other way.

[00:18:41] Stephen Manley: Yeah. You know, what I see an increasing number of people doing is, is, you know, sometimes turning off their really high end expensive everything’s duplicated DR systems that they had for let’s say 10% of their data. Maybe they, they keep it for the, the elite 5%. they take the money. They save on the other 5% and they can actually basically do DR with, with Druva for almost the rest of for everything.

[00:19:09] W. Curtis Preston: Exactly. Yeah. So keeping that, that like the ha style, right. For the, for the 1%. Right. And then, yeah, I like, I like that a lot. Right. Bringing DR to the rest of the data center instead of, to just like you said, 10% of the, um, so you, you could either save money or you could increase service and keep your costs the same. Right. But only because we’re saving

[00:19:36] Stephen Manley: That’s, that’s, that’s a hundred percent and, and, and, and I, and I think you. Again, this, this ties back to that point that, uh, you know, we, we did a, um, an event in, uh, in Texas, uh, a bunch of CIOs and, and it was interesting because when I went there, I expected that the main topic conversation was gonna be ransomware because everywhere you go, it’s ransomware.

Um, and given the economy was starting to slow, I thought, oh, okay. Maybe it’ll be about TCO. It turns out it was about talent. Each one of these CIOs, each one of these organizations basically said, I don’t have enough smart people who can do the projects that need to be done. Uh, and, and so the, the, the discussion amongst themselves was about how are you all getting around this?

Because it’s not like, you know, we’re all in the same area. We’re none of you have imaginary talent or, or, and, and so a lot of the focus was, um, how do we take our people? You know, off of, off of what are sort of more mundane tasks and free them up, train them up and, and get them to do bigger things and, and upskilling their work base, uh, was critical.

And so, so many of them are saying it’s about, can I automate and simplify the things that I’m doing? Um, and, and it used to be, you know, back back in the early two thousands, when I would hear that conversation, the implied discussion was. I’m kind of laying people off and because there isn’t the talent, it’s not a, it’s not that, that anymore.

And it really isn’t, I’m trying to lay people off to save money. It’s I need to be able to get more out of the team I have. And, and so I think that’s more and more of the conversation I’m having with people is whether it’s DR or it’s ransomware, or it’s just straight back up. Let us take care of that for you, right?

Just like Microsoft 365 takes care of your exchange. Let’s let you focus on the, the bigger. Business initiatives.

[00:21:31] W. Curtis Preston: Yeah, I, I was gonna make the same, uh, analogy, right? Nobody wants to administer Exchange, and nobody wants to administer a backup system. Right. so you’ve given, you’ve given your exchange to Microsoft 365, give your backups to us. And, and then by doing that, you free up your top talent and, and you get all of those best practices and all of those things that your top talent would have been doing for you, but without having to do, um, any of that effort.

I like that. I like that a lot. I think we’ve covered the, the subject of TCO pretty well today.

[00:22:06] Stephen Manley: I think so. And, and again, I, I, the thing I always want to close with is, you know, if, if, if someone is coming to you with a TCO argument and it’s got a three year horizon on it, you know, call them on it. If, if, if, if you can’t show a one year turnaround in TCO, then there’s a whole bunch of trust me. It’s going to be fine.

Don’t trust them. Right. So expect that one year return on investment that should get bigger, but there should be a one year ROI. Otherwise they’re full of it.

[00:22:39] W. Curtis Preston: Yeah. And, and my final comment would be that your backup system is way more complex than you think it is. And, um, You can save a lot of money by getting, by getting rid of that complexity, both in terms of things you have to buy and things that your people have to do, so well with that. Uh, thanks Stephen for chatting about TCO and, uh, thanks to, uh, our listeners.

Remember to subscribe so that you never miss an episode and remember here at Druva there’s no hardware required.