 Gweld i'n gofyn i'n argymwysio'r ysgolwyd yma. Mae'n amlwg Tom West. Rhywbeth a'r argynteg cyflawn o'r cyflawn ar y cyfrifiad i'n ddiddordeb cyflawn o'r cyflawn cyfrifiad. Rhywbeth yma, rydyn ni'n gweithio'n ymddiol ac mae'n gweithio'r ysgolwyd. Rydyn ni'n gweithio'n gweithio'n gweithio. Rydyn ni'n gweithio'r ysgolwyd, I initiate, now I typically develop on a platform called, Splunk. Is there anyone that's used Splunk or got Splunk? Anything? Yex perks. Anybody want to use Datadog or ELK? Or anything on those lines? Nope. Excellent. You can't see where I'm going to go with this. Along with working, I've gone and developed... 13 or so apps at the moment, over those 1600 downloads. The main one that I've gone and the most popular one fel gyngor. Wrth gyd am gyflawn o'r gaelio. Mae'r gaelio chi gaelio'r plwn yn eu wahanol, iddyn nhw'n gwybod gyda'r gweithio isio i gaelio a adrylu'r corfodol iddyn nhw o'r ddaf i'n gwybod gydag weithio. Felly, mae'n gwybod i'r tynnu'r bwysig cofair yn cyffredin gan gyrs. Mae'r wahanol ffrind MAC, Fyrddiau gyda'u proffesiwr professionol. Mae gennym wahanol o eich 10 dwylch, ac mae'r byddwn yn chwyn fawr i'r llyffaethol. So we work in healthcare, financial services, and we do some quite innovative and different things. And one thing that I want to talk to you about today is why you might want to use something like a data tool where you would typically throw your application logs for DevOps. Why am I here to talk about that? Well, typically those tools are generally available, they're generic in terms of what data you can add in there. The APIs for DevOps tools, GitLabs is one of the best that I've gone and used. It's really well documented, it's easy to access the format of the responses quite easy, and you can just pull that in, it's not particularly difficult to do. And then you can start to identify patterns with your data, just like you can do with inbuilt reporting within GitLab, within any other DevOps tools that you do use. But you can then also apply that report into your application logs and security incidents to then enrich that further, and all of a sudden you've got everything all in one place, which is quite nice, and it's ecosystem agnostic. So you can, it doesn't matter whether you're using, whether you're already using GitLab for everything, whether you're trying to move over and use GitLab for everything, if you've got legacy projects we heard earlier on about how companies, like large companies, they're using GitLab for a lot of their things, but struggling to get data to merge everything into GitLab. And this would be a solution for that. So that's enough about kind of the concept. Really what I want to do today is actually show you, we've done this. So right here what you can see on the screen is something that we've developed within Splunk, but if you don't want to use Splunk, you can use exactly the same methods with whichever tools that you do use. And what we're showing here is that we've got four projects which we're currently coming in, we're tracking plan quality and commit areas of your development lifecycle, and we've got four projects, like I say, we've got three that we're actually tracking plan information, one where we're checking quality, and we've got three for commits. Now that's all kind of, you know, there's nothing particularly exciting there, until I say that commit, we're moving over our data at the moment. So we were using BitBucket, we're now using GitLab. So one of those projects on there is actually BitBucket, the other two are GitLab, but it doesn't say which one's which. It just says, like, there's a status of my repositories, my projects at that particular time, and we're even able to see a timeline of how many commits we're going on. Now this is dev data, so the timeline's not particularly interesting. If you've got it into your own organisation, I suspect that it's going to be a lot more interesting and entertaining. But then what we're able to do after we've gone and summarised it, is still produce some of the details that you get off the back of that. So we've got, so we're tracking Trello, we're tracking Sonacube, and we're tracking GitLab, all three of them are entirely different ecosystems, and there are integrations with different DevOps tool sets across them. We've got to see that already today, but we're just pulling all that data into one place and actually able to report on that. And we can see for this one, which is GitLab, we've got two commits within the time period we've gone and specified, and we can see actually an activity timeline. And this is just pulling in the activities that we've got straight out of the API, from the GitLab API, ready and available. But we're not saying this is something that you have to manage totally separate. This is just automated, basically cron jobs, calling the API, pulling the data in, and putting some visuals in front of it. The thing that you probably can't see certainly at the back is that there's a table right at the bottom there, so we can click on that table, we can click on that time chart on that graph, and it's going to go and show, right, these are all the commits, these are all the activities that happen at this particular point in time, and you can select that table, and it's not going to show you any further information within Splunk or within Elastical, whichever tool you're using, it's going to take you straight to GitLab to actually show you, because why would you do anything more? GitLab is the best place, that's what you're using for your toolset. Why would you not just go there and have a look at what it is that you want to have a look at? So it's really powerful just being able to just pull those bits of information just all into one place, and what I want to do as well is just to kind of prove that it's not just me saying something, it's just theoretical. This screenshot, and it looks very, very similar, is actually Bitbucket, so this is actually telling me, this is a totally different project, we're getting it in the same screen in the same way, but this is Bitbucket, so this is our legacy project that for some reason we can't move across, and we're just showing it in exactly the same way, exactly the same method in terms of making it look really standard, really, and then linking into that toolset as well. So it kind of eliminates that problem or helps eliminate it to some degree in a totally different way. And it means that you're not having to have as much of a, you're not feeling as though you're under pressure to get everything moved over into an ecosystem as quickly as possible. You can do it at times that are a lot easier for you, it's easier to manage that process. And the last thing, like I say, I try and do talks where I have takeaways. So in this case, check out what's already available and I would always say, whichever toolset you're using, check what's already available, I guarantee there's going to be integrations and things already available. In this case, you can, if you're wanting to take, check out those screenshots and everything in more detail, go to splunk.com and you can get a free licence. There's a website called Splunkbase, which is their app store. You can go on there right now, there are integrations for, we've gone and done them for GitLab, for Bitbucket, for Trello, for SonaCube. The app that I've gone and shown you a couple of screenshots of right now will be live in 20 minutes time, which is entirely deliberate. So, but the main thing is take advantage of the APIs. Understand what your own objectives and goals are. If you're going to write this and do this from scratch, understand what it is that you can get from the APIs in the first place. It may be that you can't do something that you really want to do, but you can get 90% of the way. And if it's something really big, then break it down into really small chunks and do it that way. Go right, okay, so I've got six, seven different projects, different tool sets that I want to integrate or just focus on the first one and get that in and make sure you're happy with that, but make sure that's the most valuable one and then see what you can do based on that and go from there. Don't just aim for the moon, just do it in steps on the way. That's just kind of basic principles. I don't really want to teach you to suck eggs. And then the last thing really is just check out, we're building integrations and tool sets and new apps all the time. Within six months, six, seven months I've been in the company, we've developed seven different apps that are actually on the App Store. Like I said, we've got over 1600 downloads now. I personally have got over 13 apps available. So check out what our latest news is on convergingdata.com. We hope to add even more data and more value later on. Certainly the next thing that we're looking to do is to add alerting back in. So if you've got security products or things like that and you've got events that come in, then being able to actually post out, or to make things like merger requests and production of pipelines, even rollback of code, that sort of stuff that we're looking to do next time and next in the pipeline. Again, this is all going to be free. We're not going to charge anyone for you to be able to download it. It's just something that we, again, as part of like an open source thing, we're just making sure that we give it back to you as something almost to thank you for me because I've learnt an awful lot from the community, so therefore I want to give back a lot to the community. This is one of the best ways in which I can go and do that. That's me. Thank you very much.