Data Points

16. A Cloud DBMS Visionary

Data Points
16. A Cloud DBMS Visionary
/

In this episode, hear from Jeff Fried — the director of product management for data platforms at InterSystems — about InterSystems being named a visionary in Gartner's first ever magic quadrant for Cloud Database Management Systems.

For more information about Data Points, visit https://datapoints.intersystems.com.

 

EPISODE TRANSCRIPT:

 

Derek Robinson 00:00:02 Welcome to Data Points, a podcast by InterSystems Learning Services. Make sure to subscribe to the podcast on your favorite podcast app. Links can be found at datapoints.intersystems.com. I'm Derek Robinson. And on this episode, I'll chat with Jeff Fried, the Director of Product Management for Data Platforms at InterSystems, about InterSystems being named a vision in Gartner's new cloud DBMS Magic Quadrant.

 

Derek Robinson 00:00:39 Welcome to Data Points by InterSystems Learning Services. Hopefully you've had a happy beginning to 2021. In this episode, our first of the new year, I'll be chatting with Jeff Fried about a recognition InterSystems recently received, and what it means from a technology standpoint. Now you may have seen in other materials that InterSystems has been named a Gartner Peer Insights Customers' Choice for Operational Database Management Systems for the second year in a row. This has been mentioned at our Global Summits and our Virtual Summits, and it's something that InterSystems is definitely proud of. While that  recognition holds a lot of weight, there's another one that's very new related to Gartner that I'll be talking to Jeff about, and that's being named a visionary in Gartner's first ever Magic Quadrant for Cloud Database Management Systems.

 

Derek Robinson 00:01:29 All right. And welcome to the podcast Jeff Fried, our Director of Product Management for Data Platforms here at InterSystems. Jeff, thanks so much for joining us.

 

Jeff Fried 00:01:36 I'm delighted, Derek. It's always good to talk to you.

 

Derek Robinson 00:01:37 Of course. So today we're talking about some topics around Gartner and the Magic Quadrant, specifically for Gartner. So over the past two years, both 2019 and 2020, InterSystems has been named a Peer Insights Customers' Choice for Operational Database Management Systems, or ODBMS. That info has been displayed a lot in our marketing a lot in our website, even in our learning materials, at conferences, everything else. But one of the new things for 2020 is that Gartner released their first ever Magic Quadrant for Cloud Database Management Systems. And InterSystems has been named a visionary in that Magic Quadrant, which is really exciting and interesting. So as the Director of Product Management for Data Platforms, can you tell us a little bit about what that recognition means from a technology standpoint for InterSystems?

 

Jeff Fried 00:02:27 Sure. I'll start just by parsing what you said, which is that Gartner as a big market analyst firm has a variety of these things. The Customers' Choice is really exactly that; it's talking to customers and getting their perspective and it's really, really gratifying to get such top marks. So that's really real-world mileage. How well does our support work? How well does the product work in production? The Magic Quadrants And the Magic Quadrant for Cloud Database Systems is the first time.  Historically they have had a different thing for operational database, if you will, transactional stuff, and one for analytic databases, which is much more the OLAP and the analysis kind of tools. And this is the first time that they're together. And it's also the first time that it's explicitly cloud. So from a technology standpoint, that's a lot, and being a visionary there is very exciting. The analytics elements that they picked up on and highlighted, include things that we've been doing for a long time: very high speed ingest, strong security, strong workload management, but they also include areas that are a big investment for us now, such as advanced analytics or machine learning.

 

Derek Robinson 00:04:13 Right, right. Nice. And so, yeah, it seems like, that crosses with a lot

of the things that we've, you know, our best foot forward in many areas when it comes to IRIS and the things that we've promoted.  

 

Jeff Fried 00:04:26 Yeah, absolutely. The way that I think about this is that this represents recognition of success in areas that we may not have been traditionally in. I mean, InterSystems has a long and illustrious history, and we're best known for these operational transactional things, which are frankly much harder because of data integrity needs than some of the analytic cases. One of the things that Gartner does that I, as a nerd, like from the numbers standpoint, is they do divide things up then into use cases. So when you're building a platform, it's hard; we're building for builders, and they can do all kinds of things with it. But in particular, along with the Magic Quadrant, there's things called critical capabilities. So there's a set of use cases. For analytics, the new area to us, there's ratings of a bunch of capabilities and how they apply to a traditional data warehouse, to a logical data warehouse, to data science and machine learning, and to operational intelligence. And that, you know, it's all designed so customers can change the weights themselves, but the areas that are new from a technology perspective, like advanced analytics and multi-cloud, and the sort of financial governance elements, gained us a pretty strong position in logical data warehouse in areas where you have data that might be spread around and you want to use it for analytic cases, as well as nearly top of class in what's called operational intelligence, which is sort of a buzz word for being aware of what's happening in your operations, being smart about it, and being able to get both insight and action from that. The other side of this as a first-time thing was the cloud. This is explicitly cloud, and in a way it's Gardener's bias that they are highlighting that so much new business is moving to the cloud and that the complete majority, if you will, is already there. We are frankly, late to the cloud compared to some players. So there was even question in Gartner's minds about whether we should be included at all, based on their perception of what we're doing in the cloud. And then when we really got into it, they were very surprised. And frankly, I was surprised at exactly how much cloud work is really happening. We have a lot of customers that are running in the cloud. We have quite a number of partners that are running SAS services. We're rolling out our own SAS services and platform services. So even though we're late to the cloud, if you will, I think we have a very strong showing for that. And that's really what gave us a visionary position.

 

Derek Robinson 00:08:13 That's great. And I mean, in Learning Services, we've been involved in some of the stuff around that cloud offering as well. And you're right. It's been the last couple of years really that we've moved to having offerings on AWS and GCP and Azure and kind of things that just three or four years ago, weren't really a part of our offering at that point. But with those things in mind, what are some of those elements you see shifting in the future that will drive us more and more to cloud-friendly technologies, and how InterSystems is prepared to keep growing in that regard and be able to maintain that leader or visionary stance in that space?

 

Jeff Fried 00:08:50 Yeah. I'd say two things come to mind. One is a continuation of what we've already been doing. So as you mentioned, we already are present in these cloud marketplaces. It turns out that we have customers on every major cloud. And there is a very strong trend towards what are called orchestrators — ways that you can set up your system — so that given the cloud where you can add machines at will, and add resources, so that the topologies dynamically change, so that you can manage them along with your applications in a single fabric, if you will. And Kubernetes is probably by far the most common orchestrator; that's now something that every cloud has their own service, you know, there's, the Amazon Kubernetes service, the Azure Kubernetes services, the Google Kubernetes engine, they all are based upon this open-source orchestrator. So containers and Kubernetes are not just fads. And we've been investing in them for a couple of years. It's still the case that many of our customers don't really understand how to work with these technologies. And there are things that we still need to learn 'cause it's working very fast, but the path there in the market is more and more of a shift towards this kind of orchestrated architecture, and to things where you don't set up how many machines you're going to have ahead of time. You say, here's the work I need done, and the system adjusts to the work. So that's one front. The market is definitely shifting in this direction. There's a lot around server lists, or what a Lambda architecture is, that you can accommodate if you are working in one of these orchestrators. The other area is around using this and other cloud technologies to do more interesting architectural things. I mean, we're doubling down on a lot of our traditional strengths — on ingest, on high availability through mirroring, on consistency, on scale-out — but the cloud offers things like object store, low-cost facilities for storing data, that may not be fast and easy to access the way that we're used to with disks and SANs, but is good for certain classes of data. So we'll be investing in that area. It also offers much quicker access to things that you could get on prem to — what's an example of that? — GPU-based servers Because GPU's are used heavily in machine learning; it's all math. And people don't tend to buy and have GPU's lying around in their data centers, but you can get them easily in the cloud. Same thing with, you know, alternate architectures, like, there's a big push towards ARM, both at Amazon and Apple. So, we'll be taking advantage of those in a way that's still portable across cloud. So you know, one of my ground rules is that we want to have the same blueprint, the same architecture running, in whichever cloud you're running in, or on-prem. Gardner in their Magic Quadrant calls this multi-cloud, meaning, you may be running in different clouds in different geographies because of the presence or the regulations. You might not want to get too invested in any cloud provider because that's part of their game. Then you have a dependency. And I think as an independent, we are not Amazon Google or Microsoft, you have the advantage that we can work with any of them.

 

Derek Robinson 00:13:37 Right. And building our platform so that it can work with any of those. And to one of your earlier points, I remember actually at a prior Global Summit, one of our colleagues, Anton, and you were doing a presentation about that very topic of a machine learning model, for instance, that you need to really run some intensive training on it that requires a lot of GPU machine power, and you're not going to go buy, you know, 20 machines to do a small amount of training that you're going to do. And so being able to deploy in cloud environments to use it as needed and be able to be efficient with your spending as a business, when you're using that type of computing resource, you know, that's an important element of it as well.

 

Jeff Fried 00:14:14 Exactly. And Anton, I think, would be a great person to have on your podcast. And he's continued on with the SageMaker integration that we showed there. And it's a great example where you may not need GPU's permanently; you may just need to rent them for an hour.

 

Derek Robinson 00:14:33 So yeah, so just a final question, kind of moving into, I think you talked about this a lot already, but, you know, you mentioned Kubernetes, you mentioned kind of our being multi-cloud. Any specific features that you have kind of on the forefront or that you see as having been recently released or coming out that really play this cloud presence specifically?

 

Jeff Fried 00:14:53 Well, I think, our InterSystems Kubernetes operator, or IKO — rolls right off your tongue — is definitely something to highlight. We've had to do a significant amount of work sort of in the basement of the system, and on the way it's configured, in order to fit into this orchestration framework. And we released this with 2020.3. I think that might be worth having another podcast about, and it's really nice to see, we have a few customers running on it now and a lot more that are headed this way. The trick with any orchestrator, it's like…it's automation. So you have to set things up ahead of time and just like, think of automated testing. It may not be quicker for you to run the test the first time, but it's much quicker the second and third and fourth time. So, only a few of our customers are really familiar with Kubernetes in the areas that we've been working with them on, but we're running with it on some of our own systems. And we have some really big customers that are already using it. So I think that's worth a highlight.

 

Derek Robinson 00:16:23 Nice. So exciting stuff. And hopefully we'll continue to position ourselves well in that cloud database management system space. So, Jeff Fried, thanks so much for joining us. We'll talk to you next time.

 

Jeff Fried 00:16:33 Great. Thank you.

 

Derek Robinson 00:16:39 So many thanks to Jeff for spending some time to talk with me about being a visionary in the cloud DBMS Magic Quadrant and some of the InterSystems technologies that we will continue to see in that effort to support the journey toward a seamless and intuitive fit in the cloud. In future episodes, I think we'll definitely explore more about things like containers and the Kubernetes operator, both foundational pieces to InterSystems cloud-native approach. But for now, that'll do it for Episode 16. We'll see you next time on Data Points.


Brought to you by InterSystems Learning Services of Data Points