Open Source Summit NA 2025 | Martin Mao, Chronosphere

Clips
News
More from Open Source Summit NA 2025

Martin Mao

CEO & Co-Founder

Chronosphere

Code, community and the future: 13 takeaways from Open Source Summit NA

Beyond uniting developers under shared ideologies, goals and pain points, Open Source Summit North America ignited conversations at the intersection of code, community and change.TheCUBE goes live with insights at Open Source Summit NA

Chronosphere redefines cloud-native observability with Logs 2.0 and real-time data control

Cloud-native observability delivers deep, real-time visibility into complex, distributed systems — accelerating the detection, root cause analysis and resolution of issues across microservices, containers and serverless environments.By helping organizations control and optimize their telemetry data, Chronosphere Inc. makes observability scalable, actionable and cost-effective with its platform’s open-source foundation being key to empowering enterprises in today’s distributed, data-intensive environments, according to Martin Mao (pictured), co-founder and chief executive officer of Chronosphere

play_circle_outline Meet Matt Mao: Chronosphere CEO Showcases Cloud-Native Observability at Open Source Summit North America 2025

play_circle_outline Enhancing Efficiency: Chronosphere Logs 2.0 Unveils Comprehensive Log Management with Metrics and Traces Integration

play_circle_outline Streamlining Observability: Reducing Complexity and Costs by Phasing Out Legacy Tools for Modern Data Challenges

Info
Transcript

Martin Mao, Chronosphere

Martin Mao

CEO & Co-Founder Chronosphere

In this exclusive interview from Open Source Summit North America 2025, Martin Mao, co-founder and CEO of Chronosphere, joins theCUBE Research’s Paul Nashawaty to unpack the latest in cloud-native observability. Mao shares details on the launch of Chronosphere Logs 2.0, a major expansion of the platform’s log management capabilities designed to help organizations control data sprawl, reduce costs and improve performance across telemetry streams.

Mao explains how Chronosphere’s approach flips the traditional model of observability by helping enterpris... Read more

explore Keep Exploring

What role and focus does Matt Mao have at Chronosphere, and what is the company's approach to observability in cloud-native environments? add

What new product is being launched that enhances log management capabilities? add

What challenges are legacy tools facing in relation to data growth in the industry? add

bolt Powered by CUBE AI

Martin Mao, Chronosphere

search

Paul Nashawaty

>> Welcome to Open Source Summit North America 2025. My name is Paul Nashawaty, covering app dev space and all things open source. I'm joined today by Martin from Chronosphere. Martin, welcome.

Martin Mao

>> Thank you for having me, Paul.

Paul Nashawaty

>> Yeah, great to have you on. Why don't you tell the audience a little bit about yourself and Chronosphere?

Martin Mao

>> Yeah, so I'm Matt Mao. I'm the CEO and co-founder of Chronosphere. We're an observability platform focused on cloud native environments in particular, so containerized environments there. I do all parts of observability, including being able to ingest and store logs, metrics, traces, and all the data sources, and provide a SaaS solution on top of that. And the reason why we're here is, we're very focused on open source standards and making sure that customers can control their own format of their data, and leveraging open source standards as much as possible in observability.

Paul Nashawaty

>> Very, very cool. I mean, it's certainly not a lack of interest in this space for sure. So, well, let's just jump right into it. Chronosphere has had a lot of excitement over the last couple of days. Why don't we start with the opening, the press release, and the information around logs? Let's start there.

Martin Mao

>> Yeah, so we are launching what we call Logs 2.0, Chronosphere Logs. So it's a brand new launch of a brand new product for us. It is end-to-end log management capability, so you can imagine our ability to ingest and store logs natively into the product, and use logs along with the other data sources like metrics and traces. And on top of that, we're also providing a set of capabilities to control log data and log data growth, and that is fairly unique in the market. We've been doing that for our other data types historically, for metrics and for traces, but now we're extending those control plane capabilities over to logs as well, and really giving customers the ability to, I would say, manage the data volume growth in logs, as well as cost, while also having a great performance and experience at the same time.

Paul Nashawaty

>> Well, that makes sense. I mean, what I'm seeing, as an analyst, I always come with numbers and such. I see that enterprise logs and log data is growing 250% year over year. What we see in theCUBE research is 68% of organizations cite ballooning telemetry data is the top inhibitor of observability effectiveness. How is this helping that?

Martin Mao

>> Yeah, so that is, I would say, the secret in the industry is that you can imagine all of the existing solutions are really motivated for people to produce as much data as possible, because the more data you produce, the more data you need to store, the more companies can charge for the solutions, right? So you can imagine, the whole industry is not really incentivized to curb that growth. And what we're seeing is exactly what you're saying, which is 250% year-on-year increase in data volumes. And you can imagine, what that directly translates to normally for most solutions out there, is a 250% increase in cost of that. And if you think about the increase in cost versus how businesses are growing, most businesses are growing 10, 20, 30% year-on-year. You can see there's a huge disconnect between those two things, right? So for us, it's really about curbing that flow, really about having a solution that tackles that particular problem, and our way of doing it, and the secret with this is, not only is the data growing, but most of this data is useless. It's not useful. But it's really hard to tell what's useless and what's useful, right? If you think about observability, you instrument it before you start to create the dashboards and the alerts, and as you're instrumenting, you don't really know what data you need and how you're going to use it. So people end up producing a lot of data, and they don't really come back and focus on what's useful or not. So for Chronosphere, and what we're doing differently with logs is, we actually provide that ability to show you what data is useful, what data is useless, and then help you optimize the data so that you only keep and pay for the valuable data. And that is our way of curbing this data volume growth problem.

Paul Nashawaty

>> Well, I mean, that makes sense. I mean, on our briefing that we had, I was really impressed with the new usage analytics piece that you had. We see in our research that seven out of 10 enterprise are actively exploring ways to reduce data. Because it's exactly to your point, they're storing a lot of data, they don't know what they need. And it's no longer about alerting, it's about understanding what it is, so you can take actionable insights. So log to metrics transformations is incredibly important, and to deal with that, and you mentioned, you touched on this a little bit, that the budget and strain of scale is also a big factor. So, let's talk about usage analysts, and talk about the feature, and why that helps eliminate this.

Martin Mao

>> It's almost the missing piece. So to your point, a lot of companies are focused on reducing that data volume. And there's a lot of products out there. In fact, we acquired a company about a year and a half ago called Calyptia, and it was a telemetry pipeline. And that to me is the hammer. It's the thing that can actually go and manipulate your logs, convert logs to metrics, apply different sampling rates, de-dupe a lot of that data stream. So there's a lot of techniques for optimizing that data, and it's almost the hammer. The piece that's missing is, people don't know how to apply it. So you have all the power of the log reduction. You don't really know how and when you should apply particular optimizations. You don't know that if you apply an optimization, is that going to impact my alerts, my dashboards? Is it going to impact the end user experience? So the usage analyzer's really the missing piece there. And what it really does is, it actually starts by analyzing how you use the data first, and then it analyzes that usage and it takes that and turns it into ways in which you can smartly apply the telemetry pipeline to reduce that. So it's almost the missing piece, and it almost closes the loop of optimization there. And what we have found is, with just a telemetry pipeline, customers do get benefits from it. They're reducing their data volumes by maybe 30 to 40%. But once you introduce a usage analyzer, that optimization percentage goes up to 80 to 90%. So it's far more efficient.

Paul Nashawaty

>> Yeah, no, very cool. And that kind of goes into the next question I have, which is, we see that 63% of IT leaders are feeling forced to trade off between observability depth and cloud spend. So they're looking at this... And also, in our research we see that there's a number of different tools that are being used, right? We find that in a recent study, we did an app dev summit recently that we had research across day zero, day one, day two, and dev sec ops, and we found that for the observability side, that 75% of respondents indicate that you use six to 15 different tools.

Martin Mao

>> Yeah.

Paul Nashawaty

>> It's just amazing how many tools. So it's not exactly the most optimized way to do this kind of observability. But when we look at legacy tools as a shortcoming, how does the Chronosphere Logs here, how does it provide a path to achieve that, understanding the signal through the noise?

Martin Mao

>> Yeah, so we talked about this a little bit earlier, where the secret in the industry is that none of the legacy tools out there and their business models, they're not really optimized to deal with this data growth. Because this data growth phenomenon is more of a recent phenomenon. It really is powered by our shift over to containers. And you can imagine our workloads now are so much more ephemeral than they were before, and it's just producing a lot more data in that particular way. And you can imagine the data growth is growing faster than overall infrastructure spend. It's growing faster than a business is growing. So you can imagine these existing solutions and the legacy solutions have been optimized for a world where data growth was not a problem. The advantage we have at Chronosphere is, we saw this problem... Myself and my co-founder used to run the observability team at Uber. We saw this problem pretty early on, and we crafted the whole product and the whole company around tackling this data growth problem there. So you can imagine as a disruptor, we're able to come in with a fundamentally different business model that's focused on only paying for data that's useful. And you can imagine with that business model, it's far more efficient at tackling this particular problem. That's on one side of it, to just deal with the data growth. The other side is, there's a lot of tool consolidation, right? Because to your point, there's six to 15 tools. That's a lot of tools out there, and you do want to consolidate as much as possible, both from a cost efficiency side, because fewer vendors to deal with, generally lower costs there. You can bundle a lot of things together. And also, effectiveness of the tool. You don't really want to jump from tool to tool. So what log management does and Logs 2.0 does for Chronosphere is, it adds that capability to the Chronosphere observability platform, so that you can do more with fewer tools, and at the same time, it's doing the consolidation of the number of tools as well.

Paul Nashawaty

>> And it's also giving you that unified view, which is kind of what you're talking about. But also, we found in that same research that 54% of organizations are looking in favor of a unified view, versus having these individual tools. So the other big factor that comes up in this conversation is prioritizing real-time observability and control over the data ingestion. And what we see is, in our research, that unified MELT stack improves the MTTR, and up to 40%, versus those individual tools. So I guess I want to talk about the next thing here, which is Chronosphere's integrated logs, metrics and events, as well as tracing, providing that faster root cause analysis, right? What can you tell me about that?

Martin Mao

>> Yeah, so the way I look at our platform is, we need the underlying stores for all the different data types that you mentioned, metrics, logs, traces, and the stores, in my opinion at least, have to be specialized, because we care about the cost efficiency, we care about the performance, we care about the scalability. So we have different backend stores for all the data on the backend. On top of those backend stores, we have the control plane, which we talked about, which is, and we do this across the board here. We analyze the usage and the value of the data, and then we optimize the data so that you have to solve less. That already helps, and you can imagine reducing the noise a little bit, because there's just less data. The only data we keep in the background, and we have you search through is the data that's useful. And then our user experience is actually built a layer on top of that. For the user experience, while we do have, there's a log explorer, there's a trace explorer, there's a metric explorer, we don't think about the experience in that particular way. We think about the experience from the end user perspective, which for an end user, you're not thinking metrics, logs and traces. You are thinking, I need a problem, I need to debug end-to-end.

Paul Nashawaty

>> Fix it, yeah.

Martin Mao

>> And what this does is for us, we design the user experience with that in mind, but the more data you can feed it, the better that experience becomes. So for example, one of our user experience features is Chronosphere Lens, which is an automated user experience built for service owners and application owners. Now, if you have metric data, it shows a lot of information there, and if you add trace data to that, it shows all your dependencies and even more information. If you show add log data to that, it powers that feature even more. So the way we think about our user experience is, it's one unified user experience. It's not optimized for any data type, but the more data types you can feed it, the better that experience becomes. So it is unified, but it's more than just sticking four user experiences together. A lot of the solutions out there are, they have four separate UIs, and they're in one place, but they're really just right next to each other.

Paul Nashawaty

>> Four on a dashboard.

Martin Mao

>> And it's better than four separate tools, for sure, but I would say a truly unified experience should be thinking about from the end user perspective, as opposed to the data type perspective, and approaching it that way. And that's the way we've been doing it. And what this launch allows us to do is expand all those capabilities into Logs as well, so all of the end user features, it's not that they're just new, they just get enhanced even further by Logs as well.

Paul Nashawaty

>> No, that makes sense. And as we kind of talk about all the pieces there, about having that enhanced view and having that unified view and having the real-time access, this is requiring a lot of organizations, when I think about where we're at here in the open source community, using the tools and kind of cobbling things together is one approach. And you can do a DIY approach, it's just the devils are in the details, right? And making sure you understand it all. But more importantly, it's about uptime. And when we're looking at these enterprise apps, and they require this uptime observability to be 99% or more, right?

Martin Mao

>> Yeah, for sure.

Paul Nashawaty

>> Can you touch a little bit about how Chronosphere does that?

Martin Mao

>> Yeah. Reliability in uptime is super critical in observability in particular, because if you think about it, if you have a product or service and you are giving your end customers a three-nine SLA, like you are promising as a company a three-nine SLA, imagine observability is the thing that's measuring that uptime and measuring how up you are. So that almost has to have a higher SLA standard than what you are promising, right? So often, our customers looking at beyond three nines of SLA. And one of the other interesting things about observability is, especially as you talked about, you can put together open source pieces and run it yourself, but often that's running in your own infrastructure. There's a huge issue there, because observability is the product that is giving you insight into whether that infrastructure is up or down. So you can imagine, if your cloud provider goes down, observability is the thing that's running inside that cloud provider, in your own environment. The thing that's meant to tell you you're down is down at the same time.

Paul Nashawaty

>> Is down, right.

Martin Mao

>> So one of the things we do is, it is a SaaS solution, so it's outside of your environment, because we think it's really important to separate the concerns there. But not only that, we are a single tenanted, so we ensure all of our customers are on their own isolated tenants. That's better for reliability from one part, but it also allows us to ensure that we can run a customer's observability stack not on any shared infrastructure. So we can ensure that we're not in the same regions, and they're ideally the same cloud providers, but at least in the same regions as a customer is, so that if US east one of AWS goes down, I have a pretty high likelihood that observability will either be up, or your product will either be up, and there's no common concerns there. So it is one of the things that we do to really ensure a best-in-industry uptime and reliability.

Paul Nashawaty

>> Well, that's just one of the things. I know there's a number of differentiators that we talked through the session, but I think that's really important. All right, so let's talk about this. We're here at Open Source Summit, and you just had this great launch, and congratulations on that.

Martin Mao

>> Thank you.

Paul Nashawaty

>> That's awesome. And the team has been doing great work. I've been inundated with a lot of information, so it's awesome. So we're here a year from now, and we're talking, next Open Source Summit. What's next for Chronosphere?

Martin Mao

>> Yeah, I think I'd probably look at it from two perspectives. I do think there will be even more. And when I think about open source, we talked about, there's a lot of tools out there, but I think the best part of the open source movement is consolidation on protocols. Because if OpenTelemetry, as a particular example, that as a industry-centered protocol is so powerful, because it allows any company to control the production of their own data and switch between any vendor products that they want. So I do think on that front, there's going to be even more over the next year in terms of protocol consolidation. I'd love to see it more go towards not just on the ingestion side, but how we query the data and whatnot. I think that's definitely one path that we're going to see hopefully some development in the next year. And of course, everything we do at Chronosphere is open source protocols first, so we're very excited about that. The second topic, and we haven't talked about this very much today, but AI is obviously top of mind. I do think that over time, there are going to be more AI-native workloads, or AI workloads on top of, and I don't think these replace the compute workloads or the containerized workloads that we deal with today, but I think these are just going to be net new use cases. You can imagine a lot more companies are going to be leveraging GPUs and GPU infrastructure. A lot more companies are going to be leveraging the foundational LLMs and deploying agents. So do think there will be a lot more development in that area, and I think that's another space where you can imagine open source standards and protocols are going to be super critical there, because it's going to be the same thing where I think, for a end user, you want to make a bet on a technology choice that is going to be future-proof. And ideally that starts with the protocols there, and compatibility, and then you can switch your backend tools as you want beyond that.

Paul Nashawaty

>> Certainly exciting times ahead, right?

Martin Mao

>> Yeah.

Paul Nashawaty

>> I mean, there's definitely a lot happening and there's a lot going here. I want to congratulate you again on this launch and the success of the team. It's been awesome. Great to have you on.

Martin Mao

>> Of course.

Paul Nashawaty

>> Your insights are always impactful, they're really great. So I also want to thank the audience here, and my name is Paul Nashawaty. I'm coming to you live from the show floor at Open Source Summit North America 2025. Thank you for watching theCUBE, the leading source in tech news.