 historically, there's been this approach of taking as much data as you can, you know, I remember back when we were first doing sys logging, we would just log everything, and then we would compress it, we'd rotate it, and then if we ever needed to find anything, good luck. It would be really hard to to find this historical data. So just remember that, you know, having access to the right data that tells you what you need to know is is critical. Welcome to brand new episode of T3M, a topic of this month. And the topic of this month is observability. And today we have with us on the show William Collins, principal cloud architect at Alcubra. William, it's great to have you on the show. Great to be here. This is the first time we are talking. So I would love to know a bit about the company. So what do you folks do? So when you think about the rise of cloud adoption, so it used to be all the technology, all the intellectual property of a company, you know, lived within the four walls of the data center. And, you know, that was all the infrastructure that practitioners had to worry about. So cloud came along, you know, it really exploded. And all the application services and things that enterprise would use have gone beyond the data center now. So you have multiple clouds, multiple services. So with this proliferation of different design patterns and services just everywhere, so is the network, you know, so that plumbing, that piping, that interconnectivity, you know, bringing all these things together and maintaining control, maintaining security, maintaining performance is paramount to making things, you know, making things operational for some of these big companies. So what we do is you can think of us as a kind of like cloud networking as a service. So we provide a single interaction, you know, a single network to, you know, enable enterprises to interconnect all these things. Of course, the theme this month is observability. But when we look at observability, it serves a lot of different, you know, functions of team within organizations. Of course, it plays a very big role when it comes to networking as well. I mean, if you just look at security, we use Docker Zero Trust Network and all those things. So I want to understand from your perspective, how do you see observability and how you have seen its evolution over a year? Sure. So just kind of like I was saying, you know, everything started out in the data center when I first started working in this industry. There wasn't really such a term as observability that was thrown around, you know, we monitored things. And monitoring was pretty limited, you know, back in the late 90s, early 2000s, you know, when I really started getting into the technology landscape. So you primarily used, you know, something called SNMP for monitoring your servers, your network hardware, all of your, you know, what you would use to run your applications in the data center. So you fast forward to now, and you fast forward to like in the past, we would say, okay, is a service up? Is it down? Is it degradated? That's kind of the extent of what we were looking for. To nowadays, you have, you know, a lot more virtualization, you have clouds, every, you know, a lot of things are API driven. So API driven at the core, a lot of API transactional stuff. And you're grabbing more data, there's been an explosion of data. So yeah, it's evolved. And, you know, really to keep things simple, you could think of it like this, you have gasoline, you know, combustion engines, gasoline powered vehicles, and you have electric vehicles, right? And the tooling that you would use to interact with these vehicles is going to be different. If something breaks, it's going to be different, you know, because they're built differently, using different materials and different, you know, parts and designs. And also the data that you can get from, you know, a gasoline vehicle versus an electric vehicle is going to be different, you get a lot more from the electric powered vehicles. It'll tell you a lot more and it'll help you predict, Hey, is something going south? Hey, this doesn't look too good, you might want to get it looked at so you can get out in front of those issues. And there is, you know, kind of the way that this technology has evolved as well. Talk about the rule or importance of observability for within an organization and also you can look at it from the networking perspective as well. So the role of observability, like nowadays, if you look at an enterprise, there, when I when I talk to a lot of big companies, I hear usually three main things. So first of all, just putting it into context, you have technologies being deployed a lot more rapidly, you know, more changes are being made, there's more out there, and it's in many different places. So the first thing an organization usually wants to do is they want to know, okay, what's out there? You know, what do we have? And then they want to do, you know, some sort of health check of what they have, you know, starting out with, okay, you know, back to the green, red, yellow. So green, is it up? Red, is it down? Yellow, is it degraded? Okay, is the health good? And then beyond the health, you know, given this microservices and all these different software design patterns, these different clouds, and then also having, you know, legacy stuff still in data centers, causation is huge. So if, you know, if something is going south or performance is degraded or something goes down, understanding the causation, understanding how things are interconnected. So then you can go into, you know, you have a tough conversation with leadership to go through root cause analysis. You want to be prepared for that. You want to know what went wrong. So that way you can arm yourself in the future for making suggestions of how we can make it better. You know, and that's with networking specifically. So networking has been, I think out of all the technical verticals, probably the slowest one it seems like to evolve sometimes. We've been stuck with boxes. You know, we've been stuck with hub and spoke architectures. We've been stuck with SNMP, you know, vendor produced MIBS like management information bases to do monitoring with SNMP pulling SNMP traps. So, you know, going into the future of networking, you know, it needs to be rethought. There needs to be some, you know, a major revamp of how we approach networking and how we approach observing networking, which is what everything is built on. Now, if you look at, as you're explaining, if you look at observatory networking, networking traditionally or even today, you know, was seen as a silo because the folks were specialized in that. How much cultural change you think is needed to embrace practices like observatory or to have, you know, once again, the holistic approach towards network or you think that organizations are doing everything that is needed for them to do or you need, you feel that no, they do need some cultural change within organizations. Yeah, so culture, culture is a gigantic part of this. And I think one thing I see a lot, like historically is, you know, especially in the enterprise space, there's, you know, call it like the golden list of vendors that an enterprise is going to use. Like, I don't know who said it, but someone once said, okay, nobody got fired for using Cisco or for buying Cisco. So what happens then when you have this, you know, all these new features from cloud, all this new innovation that's happening in technology, what do you do then if one of your existing vendors doesn't, you know, have something that, you know, is going to help you solve that problem. So part of it is understanding that, you know, culture has to change with the technology. You know, it's, if you think of it in terms of like change management, you know, historically, we've had change advisory boards, and we have all this scrutiny to go over every single change. And, you know, a lot of times the folks that are validating these changes are not even technical in nature. They don't, they don't actually understand networking or these individual things. So you fast forward to now, you have CIC pipelines, you have automation, you have, you know, DevOps sort of taking over. So all of these things feed into observability too, because once you deploy this infrastructure, your operations teams, they're going to be measured on the success of how quickly they can triage, how quickly they can, you know, determine, okay, we have a problem. Do we escalate internally within our organization? Do we call a vendor? You know, how do we understand what the problem is? How do we understand what the impact is? You know, so extending our observability is going to help operationally, which, you know, feeds into everything else. When you look at cloud native, Kubernetes word, things are complicated, things get overwhelming very quickly and observability, you know, adds one more layer of complexity. Can you talk about importance of observability for organizations today? Going back, you know, again, back when you just had monitoring, you had up downs and okay, something broke, we know it's broke. And you know, it's very limited to now, organizations have to put together like a full fledged strategy, like they have different layers of this observability problem to solve. So say cloud providers, for instance, like you have observability to the cloud platform, like at the platform level, like the foundational services level, but then you go up a layer and you have observability and, you know, visibility needs for the network, for security, for different teams that are pulling in all this data. And hey, look, you could have all the data in the world, your fingertips, and unless you know what to look for, unless you're able to make sense of it, you know, you're going to be lost. And then so going further up the stack, you know, getting up to the Kubernetes and the service mesh landscape, again, that's a different set of tools, like all the tracing involved, you know, debugging between services, it's pretty complicated. So you have all these different layers. And a lot of them are dependent on the layer below them being successful. So core networking, is it the base? And that has to be successful for the infrastructure on top of that networking to be successful. And that infrastructure needs to be successful in order for the the Kubernetes and the higher level applications and services infrastructure to be successful. Of course, we talked about, you know, observability, let's talk about how are you folks helping, you know, organizations in their journey of, you know, embracing some of these observability practices? As far as just thinking about the gaps that we talked about in, you know, traditional networking as a whole, the way that we're really trying to help is, is we connect things to our platform. You know, the first thing, you know, going back to that, that idea of organizations want to know what's out there. They want to know what they're dealing with. They want to know, okay, do we have internet gateways that we don't know about? Or, you know, do we have some shadow IT going on? Do we have a lot of VMs with public IPs? You know, could data possibly, you know, is it being exfiltrated because of some of this stuff? So the first thing that we do, if we really want to give you that visibility for what's out there. And then the next thing that we want to do is help out with, okay, what is the health of the things that you're connecting to the network? You know, is it up? Is it down? Is it degradated? You know, if you want a quick answer to the health, you know, provide that health. And then beyond that, we do a lot of different just rich telemetry, you know, looking at flows, you know, looking deeper into the network and working on that causation piece of it. And we do, you know, so we do a lot of that stuff that we've built. We don't want to just take everything that we possibly can from the network and then dump it on the customer and say, here you go. Have a nice day. And we've worked with our customers to understand, like, what is actually valuable to them? What is important? Like, what do they want to see? But beyond that, we also integrate with a lot of observability platforms, like say Splunk, for instance. You know, so if an organization is made a decision that they're going to use Splunk for, you know, observability operationally, you know, we want to integrate with that and we want to provide our customers with the best of breed integrations to help them in their operational process. And then we have an open API so customers can, you know, leverage REST and pull and use data as they see fit as well. What advice you would give to organizations as they, you know, kind of build their observability journey or, you know, practice it internally? That's a good one. I would say that, you know, I know I talked a little bit about operations earlier, but we don't give, we don't give operational teams enough credit. They have hard jobs working in operations. I worked in operations a long time ago. It was probably one of the hardest jobs I've ever had. They're the first teams that get thrown in front of the trolley if something goes bad. And setting your operations teams up for success is really important. So when you think about, okay, you're building this observability strategy, you want to pick and choose vendors and platforms and things that fit to your culture and fit to, okay, how much talent do you have on staff? Like do you have, you know, do you have a lot of IT teams or, you know, do you have budget to hire more and to train up and to do more, you know, think of it as like do it yourself and build things out internally? Or are you a little bit lean? So the leaner you get with staffing, especially now in this market that we're in, you know, looking at platforms that abstract and take a lot of those burdens away from your operations, minimize the touch points, minimize the interaction surfaces to help you be successful, I think is important. So just limiting the scope of what you have to worry about as an organization is key. And just always remember never, you know, I think historically there's been this approach of taking as much data as you can. You know, I remember back when we were first doing sys logging, we would just log everything. And then we would compress it, we'd rotate it. And then if we ever needed to find anything, good luck. It would be really hard to find this historical data. So just remember that, you know, having access to the right data that tells you what you need to know is critical. William, thank you so much for taking time out today and of course talk about the company and also observability. Thanks for all those insights. And I would love to chat with you again. Thank you. Thank you.