Jonathan Symonds, MinIO | KubeCon + CloudNativeCon Europe 2022

This is digital anarchist. Hey, welcome back. Good to have you back with us here at kubecon and Cloud nativecon.

I'm at the pleasure being joined by Jonathan Simons. Who's CMO with minayo neighbors here in our in our booth alley. Welcome Jonathan.

Yeah, I've watched you interview the the intelligence of the cloudnative. Okay A little bit of a intelligence gathering. Hey what tell us about yourself and tell us about Min IO and we'll kind of dive into the data world and kubernetes.

Yeah. Yeah. So my name is John Simons.

I am the CMO of men. I always you mentioned I was the first marketing hire. I've been there a little over three and a half years now and it's been a really extraordinary Journey because the world has changed a lot from a technology perspective in that time men.

IO is is an object Storage Company. We're known for a couple things one is that we're multi-cloud. So we are on every public Cloud.

We are on every private kubernetes distribution open shift 10 to Tuesday. Ranch or HB. We're on bare metal.

We're at the edge. The binary is only a hundred MB. So you can have the same consistency from The Edge.

So the core and we're also known for performance, which I think will probably talk about here a little bit but you know object stores are now really primary storage in the cloud and so they have to be fast and so they have to be able to run databases that we run Ai and ML workloads and so forth. And so those two things really are the are the primary thing as things that distinguish nio from the rest of the rest of the world of object storage. This seems like object storage speaks directly to developer and kind of the cloud native architecture and as well as kubernetes is approach to storage in general so clearly it's it sounds like you're you're dressing that but even more so is the data sprawl that's across multi-cloud non-cloud environments that you'll run in lots of different locations.

So that's part of the problem. You're addressing correct? Yeah.

So listen multi-cloud is one of the great challenges and Enterprise. Today because whether it's by acquisition, whether it's by Shadow it or whether it's by Design or whether it's you know, any number of reasons all large Enterprises are multi-cloud Enterprises. Nobody is in a single cloud and if you are in a single Cloud you basically have taken on risk that you probably want to start to mitigate at this point because so much of your business may be dependent upon that single cloud and so multi cloud is a great Challenge and the way it is solved is through kubernetes and so men IO is a kubernetes native Object Store.

The company was started in 2014 containerization was in full swing orchestration still wasn't fully decided right you had never really and yeah. Yeah and you had kubernetes obviously, we all know that kubernetes ran the table and extinguished everybody else from from that market, but that was always in our DNA so containerization orchestration and then kubernetes was a natural for us. That's because Restful apis you talk about developers.

That's what they want. They want https. They want to restful apis.

And so that's what that delivered them. And that's what that's three delivered them as well. And so the ability to basically take S3 and through kubernetes make object storage available anywhere is what we're providing and that's what Enterprises are Desiring today.

AWS ss3 is a remarkable product. We owe them a huge debt of gratitude having said that it only runs inside of AWS ss3 data to your point exists everywhere and is being created everywhere from The Edge to you know, the core it's just being created in all these places increasingly by machines. And so what you need is the ability to run that AWS like infrastructure anywhere you want to kubernetes gives us the ability to deliver that for Enterprises and it gives you that consistent interface pattern storage.

Across all those diverse environments so you don't have to worry about API differences between S3 and a Google cloud or whatever. It might be that our San or wherever it might be. Right.

Yeah, and that's a really good point because you know there are you have the blob API right in Azure, you've got Google's got its own cloud Object Store API, and then there's S3, and those three don't talk to each other right? It's a huge issue. But if you can run men IO across all three of those things you solve that problem instantaneously and you can just write your applications which are probably already written for restful apis and S3 today and you can run it across that and that goes for Alibaba cloud and that goes through IBM cloud and Oracle cloud and any Cloud you can think of so it is this ability to have a consistent API consistent performance consistent experience across all of those different platforms and again the cloud and we talk about it in terms of when we say it we often say oh, you know, it's the public Cloud but the cloud is really a mentality.

It's an operating model and That's what I think one of the big themes that we're seeing here at this kubernetes is that it's it is an operating model and whether it's location is independent of that operating model. And so, you know having that cloud approach is what's driving adoption from nio. Let's talk about you know, one of the things that always comes up with storage is performance.

Yeah, and and thinking about object. I mean there was a day when object debate databases were new things what they weren't high performance yet. And you know as we kind of folded that in more to a software architecture than thinking of it as a database, what does that world look like what's performance look like in an object storage world today?

Yeah. So I think historically people looked at object storage is sort of cheap deep and slow and that was a function of what the appliance vendors we're really designing for which was archival Data S3 changed the game on that. If you go back and look even at the early days of S3, it's a very fast service and that allowed object storage to become the primary storage in the cloud.

Redshift is on top of object storage snowflake is built on top of babish stories bigquery is build on top of object stores. All the most successful services that run in the cloud are object storage. They're based on object storage.

That's because object storage was basically made fast Mido followed in those footsteps. We've invested very very heavily in optimizing for ABX 512 chip sets for you know, our friends at Intel or right next to Warriors. And so what we that goes into a racial coding that goes into encryption that goes into everything else and so we're producing benchmarks that are let 365 gigs a second on reads 160 on rights on 32 nodes of ndme.

Right? So at that speed any workload now comes into play so databases can speak directly to the objects to where inquiry the object store directly and you're gonna see more more announcements in the coming weeks and months of first class giantly deployed. Across the globe databases that are now looking and making object storage their first class citizen from a storage perspective.

So that's what performance brings it brings database workloads the AI and ml workloads were already on object storage because of the requirements for both performance and scale and performance at Scales a totally different animal and kubeflow ships. For example with mid IO embedded in it, but you see those types of workloads and that's what you get out of performance. It's not just vanity numbers.

It's about business results for customers. It's interesting you had database products Technologies to the equation. Now, you're open up to a much wider set of applications that is accessible to the cloud.

Yeah. Yeah, and I think that's very uncomfortable for the traditional sand and Nas folks because they they fight this concept that object storage primary storage, even though the evidence in the cloud is clearly there. They would like you to believe that again, it's it's slow and but that's it's just Much more efficient and elegant way to approach the cloud and it's and it is kubernetes native.

But I say, yes, it's very much in the cougar and heavy kubernetes architecture and very comfortable for the developers don't have to adopt it. They're already doing it. So they think of that or they they know it.

It's just this is just in their DNA and you know, one of the things that why we love coming to kubecon is these are our people like every single person, you know, several interviews like t-shirts and backpacks. This is my this is my people. Yeah, but you know everybody who comes up like we love me now, we're running manio and it's just it's a everybody we've interacted with they know us already.

We're not having conversations about what is object storage or what is mean I know it is, you know, we just love what you're doing. Please keep doing it. Now you you have I think enough I'd say unique but certainly a very desirable adoption model.

You don't have a big sales force. Yeah, you customers. I hope I say come to you.

Maybe that's that's overstating. I Maybe not it tells about how people get to you get yeah, so it is it is very we've built a different machine. It's a much more efficient machine and that's allowed us to do a lot of things differently than other organizations.

We don't have a commission sales team. We have some really smart business people who have engineering backgrounds that if you squint say they they look and feel like somebody who's a senior advisor and that's exactly what they are. But we don't have any commission sales people 100% of our business comes inbound and it literally comes inbound through Partners or it comes in down through.

Hello at ninio or sales at men IO or they walk up the booth and it's time for us to talk about a commercial license. We've gotten to this size and so it's agplv3. So it's frictionless from an adoption perspective but agplv3 is an open source for open source the applications model.

It's a purist model. But if you're building a commercial application, you're gonna need a commercial license and so it is well known and understood. And it's it's a really good model for us and it's allowed us to get to the point where I right now.

3 million Docker pulls a day and you know, we're gonna pass a billion Docker pulls here in a few months in in Aggregate and they're just extraordinary numbers from a growth perspective and just from and you're growing off of huge numbers to start with. So we feel really good about the trajectory of what what's happened in the nio these days. Well speaking of a billion you had a pretty impressive really impressive fundraise, I think around the app and back in January.

Yeah. Yeah, you know the world's changed a little bit since then, I think we all know that in terms of the great revaluation as they're calling it. But yeah, we closed 103 million dollars at a billion dollar plus valuation Intel Capital came in and was the lead investor on that, you know Pat gelsinger personally presented the term sheet.

That's a being garima. Yeah. We aligned so well with his You know view of the world these days which is developers of the value of the engine of value creation and the Enterprise he believes that deeply that comes from his VMware days, but he also believes that increasingly data is going to be created outside of the data center.

And so you have to be able to operate at the edge and then IO does that very well. So it was a very good fit. They were a previously an investor in men IO we also brought in soft Banks Vision too fun, alongside all of our existing investors, but it's exciting and you know that 103 million is really it's a marker to the rest of the industry that we're here for a long time.

We went five years on our first series a which was 23 million dollars. Trust me. We can go a long way with the efficient model that we built without having sales people and sales engineers and offices all over the globe with a hundred and three million dollars and our Revenue ramp just keeps on going up.

So it's it's super exciting. Yeah, I don't have dated back. Up, but I think Pat's time at I gelsinger this time at VMware really reshaped his strategy for going back to Intel and helping Intel think in a software world what Intel's role is and how to how to work with companies like yourself that are you know, let's face it.

You're not a nest. You're not you're not Hardware storage. You are object storage to think about it much differently.

Yeah. I think he he understands. You know, what software to find storage is why it's important in the marketplace how that has to be optimized on Intel chipsets.

I don't think there is any better person for that job at this time. And I think that's one of the reasons that he went back. I think he knows a very strong moral obligation to the Intel brand and I think he probably recognized that if it wasn't him, you know, it may go in a different direction and you know, I think you think about Mission driven individuals.

I think he's really qualifies in that regard and like I said, I think you can already See his his imprint on how they are operating today, and I think it's exciting and it's good for the rest of the tech industry. Yeah investment in Minnesota is a good example of that too. Yeah.

Yeah vagina. It's been fantastic talking with you appreciate the good neighborhood neighbor without fences, you know, we can actually the old saying goes so be sure and check out man. I always as we talked about it's really an inbound model and a lot of people go check it out and get started and you know connect with with what you can do with object storage in kubernetes.

So wish you the best have a good show. Thank you very much joining us. I appreciate it.

I told you we have great interviews and I delivered so come back. We've got some more lined up for you that are great like Jonathan. So thanks for joining us.

We'll see you in a few.