 Good evening, welcome back to theCUBE. Live in Los Angeles, we are at KubeCon Club, NativeCon 2021. Lisa Martin with Dave Nicholson. Rounding out our day, we're going to introduce you to a new company, a new company that's new to us. I should say LogDNA. Peter Cho joins us, the VP of product. Peter, welcome to the program. Thanks for having me. Talk to us about LogDNA. Who are you guys, what do you do? So, you know, LogDNA is a log management platform. Traditionally, we've been focused on, you know, offering log analysis, log management capabilities to DevOps teams. So your classic kind of troubleshooting, debugging, getting into your systems. More recently, maybe in like the last year or so, we've been focused on a lot of control functionality around log management. So what I mean by that is a lot of people typically think of kind of the analysis with the dashboards, but with the pandemic, we notice that you see this kind of surge of data because all of the services are being used, but you also see a downward pressure on cost, right? Because all of a sudden you don't want to be spending two acts on those digital experiences. So we've been focused really on kind of tamping down kind of controls on the volume of log data coming in and making sure that they have a higher kind of signal to noise ratio. And then, you know, I'll talk about it in a little bit, but we've really been honing in on how can we take those capabilities and kind of form them more in a pipeline. So log management, DevOps, you know, focusing on log data, but moving forward, really focus on that flow of data. So when you talk about the flow of data and logs that are being read, make this a little more real. Bring it up just a level in terms of data from what? What kinds of logs, what things are generating logs, what's the relevant information that's being kept track of? Yeah, I mean, so from our perspective, we're actually agnostic to data source. So we have assist log integration, we have kind of basic APIs, we have agents for any sort of operating system. Funny enough, people actually use those agents to install log DNA on robots, right? And so we have a customer, they're one of the largest e-commerce platforms in the world, and they have a warehouse robots division, and they install the agent on every single one of those robots, they're running like ARM64 processors, and they'll send the log data directly to us, right? So to us, it's no different, a robot is no different from a server, it's no different from an application, it's no different from a router. We take in all that data. Traditionally though, to answer your question, I guess in the simplest way, mostly applications, servers, firewalls, all the traditional stuff you'd expect kind of going into a log platform. So you mentioned a big name customer, I've got to guess as to who that is, I won't say, but talk to us about the observability pipeline. What is that, what are the benefits in it for customers? Sure, so like if we zoom out again, you think about logs, traditionally, I think a lot of folks say, okay, we'll ingest the logs, we'll analyze them. What we noticed is that there's a lot of value in the step before that. So I think in the earlier days, it was really novel to say, hey, we're going to get logs, we're going to put it into a system, we're going to analyze it, we're going to centralize it. And that had its merits, but I think over time it got a little chaotic and so you saw a lot of the vendors over the last three years consolidating and doing more of a single pane of glass, all the pillars of observability and whatnot. But then the downside of that is you're seeing a lot of the teams that are using that then saying being constrained by single vendor for all the ways that you can access that data. So we decided that the control point being on the analysis side, on the very far right side was constricting, so we said, okay, let's move the control point up into a pipeline where the logs are coming to a single point of ingress. And then what we'll do is we will offer views, but also allow you to stream in other systems. So we'll allow you to stream into a SIM or a data warehouse or something like that, right? So we're still trying to nail down the messaging. I'm sure our marketing person's going to roast me after this, but the simplest way to think of an observability pipeline is it's the step before the analysis part that kind of ingests, processes, and routes the data. Yeah, this is theCUBE, by the way. Neither one of us is a weather reporter. So we're, so the technical stuff is good with us. Yes, it is. What are, and talk to us about some of the key features and capabilities and maybe anything that's newly announced or going to be announced. Yeah, for sure. So what we recently announced early access on is our streaming capabilities. So it's something that we built in conjunction with IBM and with a couple, you know, large major institutions that we were working with on the IBM cloud. And basically we realized, as we were ingesting a lot of data, some of those consumers wanted to access subsets of that data in other systems, such as Q-Radar or, you know, a security product. So we ended up taking, we filter down a subset of that data and we stream it out into those systems. And so we're taking those capabilities and then bringing it into our direct product, you know, whatever you access via login.com. That is what's essentially going to be the seed for the kind of observability pipeline moving forward. So when you start thinking about it, all of the stuff that I mentioned where we say we're focusing on control, like allowing you to exclude logs, allowing you to transform logs, you take those processing capabilities, you take the streaming capabilities, you put them together and all of a sudden that's the pipeline, right? So that's the biggest focus for us now and then we also have supporting features such as, you know, control APIs. We have index rate alerting so that you can get notified if you see aberrations in the amount of flow of data. We have things like variable retention. So when a certain subset of logs come in, if you want to store it for seven days or 30 days, you can go ahead and do that because we know that a large block of logs is going to have many different use cases and many different associated values, right? So let's pretend for a moment that a user, someone who has spent their money on log DNA. Is putting together a Yelp review and they've given you five stars. What do they say about log DNA? Why did they give you that five star rating? Yeah, absolutely. I think, you know, the most common one, and it's funny is Yelp because we actually religiously mine our G2 crowd reviews. And so the thing that we hear most often is ease of use, right? A lot of these tools, I mean, I'm sure, you know, you're talking to founders and product leaders every day. With developers, like the bar, the baseline is so low, a lot of organizations were like, we'll give them their coders, they'll figure it out, we'll just give them docs and they'll figure it out. But we went a little bit extra in terms of like, how can we smooth that experience so that when you go to your computer and you type in QPCTL, blah, blah, blah, two lines, and all of a sudden all your logs are shipping from your cluster to log DNA. So that's the constant theme for us and all of our views is, hey, I showed up, I signed up and within 30 minutes I had everything going that I needed to get. So fast time to value. Yes. Which is critical these days. Absolutely. Talk to me. So here we are at KubeCon. We've seen, CNCF community is huge. I think the number I saw yesterday was 138,000 contributors, lots of activity. Of course we're in person, which is great. We can have those hallway networking conversations that we haven't been able to have in a year and a half. What are some of the things that you guys have heard at the booth in terms of being able to engage with the community again? You know, the thing that we've heard most often is just like having a finger on the pulse. It's so hard to do that because when we're all at our computers we just go from zoom to zoom. And so unless it punches you in the face you're not aware of it, right? But when you come here, you look around, you can start to identify trends, you hear the conversations in the hallway, you see the sessions. It's just that sense of, it's almost like a phantom limb that sense of community and being kind of connected. I think that's the thing that we've heard most often that people are excited. And you know, I think a lot of us are just kind of treating this like a dry run, like we're kind of easing our way back in. And so it, you know, it's felt good to be back. Well, they've done a great job here, right? I mean, you have to show your proof of vaccination. They're doing temperature checks so you can show your clear health pass. So they're making it. We were talking to the executive director of CNCF earlier today and you're making it. It's not rocket science. We have enough data to know that this can be done carefully and safely. Don't forget the wristbands. That's right. And did you see the wristbands? Oh yeah. Yeah, it's great. I was on the fence, by the way. I was like, I was a greener yellow depending on the person. Yeah, yeah, yeah. But giving everybody the opportunity to socialize again and to have those conversations that you just can't have by Zoom because you've seen someone that jogs your memory. And also the control of do I want to shake someone's hand or do I not is they've done a great job. And I think hopefully this is a good test in the water for other organizations to learn. This can be done safely because of the community. You can't replicate that on video. Absolutely. I'll tell you this. For us, this is our event. This is the event. For us, every single year it's only an event we hear about at the end of the day. What are some of the things that you've seen in the last year in terms of, we were talking a lot about the adoption of Kubernetes, kind of where is it in its maturation state. But we've seen so much acceleration and digital transformation in the last 18 months for every industry. Businesses rapidly pivoting multiple times to try to survive one and then figure out a new way to thrive in this new, I'll call it the new now, I'll borrow that from a friend at Citrix, the new now, not the new normal, the new now. What are some of the things that you've seen in the last year and a half from your customer base in terms of, what have they been coming to you saying, help? You know, I think going back to the earlier point about time and value, that's the thing that a lot, so a lot of our customers are very big Kubernetes, they're big consumers of Kubernetes. I would say, for me when I do, we do our QBRs with our top customers, I would say 80% of them are huge Kubernetes shops, right? And the biggest bottleneck for them actually is onboarding new engineers, because a lot of them, we have a customer, we have a better mortgage, we have IBM, we have Rapi as a customer of ours, they're like the Latin American version of Instacart, they double their engineering base in 18 months, and so that's, I think it was maybe from 1500 to 3000 developers or so. So their thing is like, we need to get people onboard as soon as possible, we need to get them in these tools, getting access to their logs, to whatever they need, and so that's been the biggest thing that we've heard over and over again, is how, A, how can we hire? And then, B, when we hire them, how do we onboard them as quickly as possible so that they're ramped up and they're adding value? How do you help with that onboarding, making it faster, seamless, so that they can get value faster? So for us, we really lean in on our customer success teams, so they do trainings, they do best practices. Basically, we kind of think of ourselves, given how much Kubernetes concentration we have, we think of ourselves as cross-pollinators, so a lot of the time we'll go into those decks and we'll try to learn just as much as we're trying to teach, and then we'll go and repeat that process through every single set of our customers, so a lot of the patterns that we'll see are, well, what kinds of tools are using for orchestration, what kind of tools are you using for deployment, how are you thinking about X, Y, and Z, and then even our own SRE teams will kind of get into the mix and provide tips and feedback. Customer centricity is key. We've heard that a lot today. We hear that from a lot of companies. It's one thing to hear, it's another thing to see, and it sounds like the Yelp review that you would have given, or what you're hearing through G2 Crowd, I mean, that voice of the customer is valid, that's the only validation, I think, that really matters, because analysts are paid, but hearing that validation through the voice of the customer consistently, lets you know we're going in the right direction here. Absolutely. I think it's interesting that ease of use comes up. You wonder if those are rolling anonymous reviews. You don't necessarily associate open source community with cutting edge, it's like we're the people on the pirate ship, and so when people start to finally admit some ease of use would be nice, I think it's an indication of maturity at a certain point. It's saying, okay, not everyone is going to come in and sit behind a keyboard and program things in machine language every time we want to do some simple task. Let's automate, let's get some ease of use into this. Well, and I'll tell you, in the early days, it drove me and our CEO Tucker, it drove us nuts that people would say ease of use, so we're like, that's so shallow, it doesn't mean any of all that. However, but to your point, if we don't meet the use case, if we don't have the power behind it, the ease of use is abstracting away, it's like an iceberg, right? It's abstracting away a lot, so we can't even have the ease of use conversation unless we're able to meet the use case. So what we've been doing is digging into that more, okay, ease of use, but what were you trying to do? What is it that we enabled? Because ease of use, if it's a very shallow set of use cases, it's not as valid as ease of use for petabytes of data for an organization like IBM, right? That's a great, I'm glad that you dug into that, because ease of use is one of those things that you'll see in marketing materials, but to your point, you wanted to know what does this actually mean, what are we delivering? And now you know what you're delivering. Peter, thank you for sharing with us about LogDNA, what you guys are doing, how you're helping your community of customers and hearing the voice of the customer through G2 and others, good work. Thank you. And by the way, I'll be remiss if I don't say this. If you're interested in learning more about some of the stuff that we're working on, just go to LogDNA.com. We've got, I think we've got a banner for the Early Access programs that I mentioned earlier, so at the end of the day, to your point about customer central city, everything we prioritize is based on our customers, what they need, what they tell us about. And so, you know, whatever engagement that we get from the people at the show and prospects, like that's how we drive our road map. Yep, that's why we're all here. LogDNA.com. Peter, thank you for joining Dave and me today. We appreciate it. Thanks for having me. Our pleasure. For Dave Nicholson, I'm Lisa Martin, signing off from Los Angeles today. The cubes coverage of KubeCon, CloudNativeCon 21 continues tomorrow. We'll see you then.