Streamline Your Document Workflow with airSlate SignNow's Pipeline Tracking Tool for Quality Assurance

How to create outlook signature

hey everyone drew horn here thanks for joining us at illuminate2020 and the devops track today we're going to be talking a little bit about releasing high quality code faster through optimizing ci cd pipelines just a quick agenda today we're going to go through some intros then we're going to talk a little bit about some of the challenges and how teams are optimizing the ci cd pipeline and then we're going to jump into a gannett use case we're going to talk about the connect quality mission and then we're going to talk about how sumo logic is being used at gannet and then finally some next steps around how we're adding to the dashboards that connect are using to give them better visibility into the ci cd process so who am i ing to to sumo one of my sales peers put this lovely meme together in my my first month at sumo i am the technical business director for apps integrations and solutions at sumo logic and apparently i've got your integrations right here mike hi thank you drew my name is mike and these are the data points about me so first and foremost i am a father and this is a picture of me and my boys playing on uh climbing while i bit for them in our house to burn off some energy additionally i am coined a troublemaker and when this uh text came through the system over my one of my employers uh i laughed out loud and giggled hard because i'm going to troublemaker and being i'm okay with being a troublemaker because inside the qa industry i feel you have to continually ask the questions why to figure out the problems that we're trying to solve so i'm okay with being a troublemaker i like to play outside the rules as you can see from my younger years of playing in my toy box not storing the toys in there additionally another data point about me is that my my wife and i are licensed foster care parents and we're advocates for the utah foster care foundation and all of our children are adopted great thanks mike so i just wanted to spend a couple minutes talking about some of the challenges related to building and developing software as well as the solution that we put together here at sumo logic so i think everyone's familiar with this diagram here in the middle we're looking at the software development or devops lifecycle call it what you want you know this is a process that at a high level is very similar across all of the you know organizations and teams that are building and developing modern software however one of the challenges here is that when it comes to the actual implementation uh a lot of the different teams even at a single organization are implementing their pipelines differently and one of the challenges that that creates is pipeline diversity or pipeline sprawl looking at you know kind of the the items below the pipeline we have all of the different constituents or all the different teams that are involved in the software development life cycle and so when it comes from ideation and going all the way down to the pipeline into production uh you know there's there's several different pairs of eyes right that are looking at and touching the software as it's being uh built and tested and released and deployed and you know any friction that occurs uh at any of those stages is gonna potentially hold up the the build and test process or slow down deployments and affect your lead time overall another challenge is you know teams are are encouraged to you know use the tools that work best for them when it comes to building and and developing and deploying software and you know that's really helpful for individual teams they're using the tools that they want and they know that they know well however again when you start trying to collect data across all of the different teams and understand you know how best can we optimize our software development and delivery process processes across the organization this can become a big challenge all of these tools are emitting data of a different structure they only provide siloed kind of views of the of the full devops life cycle and so kind of summing all of that up right teams build and and deliver software differently the devops tool chain is fragmented and fraught with data silos this results in you know poor limited collaboration across teams and it also drives a lack of visibility across the devops life cycle and finally creates challenges around building standard data driven and metrics and insights to help optimize software development and delivery so talking a little bit about our solution um we have it's called software development optimization uh you know we put this solution together to try to solve these challenges right so what we're doing is we're allowing teams they can take any you know any software in the devops tool chain from the software development ecosystem and they can plug that directly into sumo logic and what we do here that's a little bit different from some of the other solutions out there is we leverage the power of our analytics platform to enrich normalize and correlate that data across the entire software development process what that allows us to do is real-time performance benchmarking of the deployment pipeline as well as end-to-end optimization so what we're doing is we're taking all of these logs that are being fed into our platform we're deriving some performance metrics that i'm going to talk about a little bit in a second that tell teams how they're performing and then we're providing some underlying metrics and actionable insights that show teams you know how are these processes contributing to positive or negative trends in their software development and delivery performance the end goal here is to provide teams this visibility so that they can make data-driven decisions around how to optimize these processes and what that correlates to what that leads to is faster release velocity improve reliability and security as well as greater team collaboration so let's talk about some of the industry leading metrics that organizations use to benchmark their software delivery performance over the past seven years dora or devops research and assessment has identified four key metrics that are tied to software delivery performance if you're an organization that is building software and is looking to get better at releasing revenue generating generating features faster and more reliably then these are the four metrics that you really want to focus on so the two on the left are measures of market agility or throughput you have deployment frequency which is how often code is deployed to production and you have lead time which is the time for merge commit to deployment uh to production then on the right you have measures of reliability you have mean time to recovery or mttr and that's how long it takes to recover from production incidents and then finally change failure rate which is the percent of deployments that subsequently require remediation these are really important metrics to monitor and with the solution here at sumo logic we make it really easy to pull in data from all of the different uh sas tools that you're using as part of your devops life cycle and aggregate all of that data into our platform so that these metrics can be computed in real time again by correlating data from across all of those tools so what i'm going to do is i'm going to head it over hand it over to mike next and have him talk through a little bit of how gannett is levering leveraging our platform to optimize their ci cd pipeline thanks drew that's a lot of information to cover and so what i'm going to be showing is some information associated with who we are and why we're involved with the software development observability process here at gannett we are the leading and local national media and market solution company we span 46 states of coverage for local media as well as the usa today network uh for the nationwide news gathering that we have we have lots of moving parts within our organization to be able to support our end users using our systems to get quality data for their news we have 2 000 plus journalists 100 plus newsrooms 150 million unique users each month we have 1.2 excuse me 1.1 billion users monthly page views and we have over 500 plus digital products that span uh the united states as well as some coverage inside of europe as part of the mission as the aspect associated with quality we want to be able to gather data to help support the dora metrics that drew just spoke about to be able to have this data globally accessible to the organization from the executive level down to the lower systems to be able to say the individual development teams the individual application teams to be able to see the state and quality of their deliverables that they're providing for the usa today network what what that provides is it provides a global accountability to the quality of what we are delivering it also improves the quality knowledge within the organization and it provides transparency to the whole organization so that we can figure out what's happening with our data and to speed up the release cycles of innovative ideas that we produce within our network so that's a coverage about our mission now what i'm going to be showing is some data points or dashboards into some of the visibilities that we have created for use within the organization at usa today the first dashboard that i want to show is it's tracking metrics associated with quality and when i say quality i mean the test results that we're actually executing and as you can see the numbers are high we have lots of data testing that we are doing associated with quality to make sure that the applications that we release are very stable and uh can meet the high demand that we have with our usership what we're tracking here is past percentage rate failure rates as well as skip counts and some of you might be asking yourselves why would you care about skips skips are important when we're looking at testing because if you're executing skip tests you are utilizing resources and spending money on test and logic that's not being used and so we look at these metrics to decide hey is this something that we should be doing and if it is let's fix it so that we can get it to pass if it's not let's delete it out of our test suite and make sure that our past percentages are improved the next dashboard that i'm going to be showing you is giving you some results associated with our build times so as part of the dora metrics we want to be able to make sure that the quality of our build states and the build times are low and so what this dashboard is showing you is spikes that we have associated with the past value rate specific to individual builds and as you can see we have lots of builds running we can also narrow the data down and filter it down into aspects of individual teams and who's doing what and where and how and why are they doing that we can check their metrics based on their past percentage rates down to the individual team and again it's not necessary to point fingers or place blame on what's going on it helps us have quality conversations with people and the teams to make sure that we can make the right decisions using quality data another view that i'm showing here is the pass or excuse me not the past but the failed percentage rate and this is important because the way we've lined this up is on the right side is the uh two-week span of a sprint and on the left side is the one week so what's happening this week and then what's happening over the whole sprint running a two-week two-week sprint cycle we categorize the data based on certain testing types that we are executing and you can see what's happening as far as percent wise in the lower two panels as i mentioned before skip tests are important and we want to analyze what's going on there and you can see the ups and downs the ebbs and flows of this data and we have that isolated down by product by test type and again this flows based on the two-week cycle of a sprint for this week and what's happened overall the next slide is one of my favorite slides this dashboard prevent presents data to a specific application there's a certain testing type that we do within the organization that does analytics testing and this is very key to the organization because we want to make sure that our analytics calls that are firing on our pages of all those pages that our end users are looking at are firing correctly if they're not firing correctly we're not getting paid correctly and we need to make sure that that's correct so this is one of my favorite dashboards that we put together to provide the insight that we need to see as part of the initiative we have going on with our organization we're transitioning some of our local media sites into a standardized production environment and that platform has to be tested to make sure that those migrations are working correctly and this dashboard has actually been key to that migration process because we can see within 15 seconds of when a deployment has happened whether or not the results are happening correctly so the way we break this stuff down is that we have a timeline of what's happening within our organization for each one of the vendors that we're looking for and then we break it down by pass failure type for page type types of vendors what platform that's going on and all the markets that we have migrated over to this new platform so this is this is uh one of the key dashboards that we use within organization and it's pretty exciting and fun to watch through watch or look through another aspect of quality is that we have a strong initiative make sure that the performance of our pages are on point and so we do some uh web page test performance testing and this dashboard is a representation of that data that's coming into the system these are fun to see because we can see when the page loads for the first time we can see how fast that's taking or how long that's taking and we can make adjustments to the page load times or the data load times behind the scenes to make sure that our performance is still on point also associated with the quality mission at the beginning that i spoke about is we want to make sure that the quality knowledge within the organization increases and what we're looking for is gaps within those knowledge and so that we can use those as training exercises to help improve development practices quality practices or just a general understanding of how we deem quality and why it's so important one of my monitors of philosophy is associated with quality is quality is a team effort and we want to make sure that everybody is focused on quality because at the end of the day we need to make sure that quality is actually out there for all of these local sites and media outlets that we have so what we're representing here is actual visibility into our repos associated with our github activity we want to see hey who has too many pr's open how long are the pirs open are they not closing prs are they doing lots of commenting on prs and then we can use this data to help educate the development teams and the quality teams to make sure that things are flowing through the system in appropriate manner now i'm going to turn the time back over to drew and he's going to talk about some of the new dashboarding that they're coming up with great thanks mike so what i want to do is talk about some of the next steps that we're doing in in partnership with gannett um what we have here are a couple of dashboards that we're working on and the first dashboard that i'm going to show here are the key four dora metrics so as we mentioned earlier we have deployment frequency on the left with with some time series and trend data same with lead time change failure rate and time to restore service what's nice about this dashboard is that we've um baked in the benchmarks from the the the 2019 state of devops report so for elite performers right you're going to get a green dashboard and then as as you're kind of you know in the high or middle or or low range of performance we're going to show that right here on the dashboard so you can drill down into some of our drill down charts and then get a better understanding of what's contributing to these metrics another way to view the data is our ci cd pipeline overview and so what we've done here is if you remember you know the slide earlier where we showed the pipeline we're providing um some high level status and and tabular data for each stage of the pipeline from right to left so we have alerts deploys builds pull requests and issues all right here and then again we're giving you a status of you know what's going on right now in the pipeline as well as you know some recent history on the last 10 failed builds or pull requests or deployments or alerts and then finally um another dashboard you know that we're wanting to provide you know for teams to give them the right observability that they need to you know drill down into the pipeline and understand what's going on in order to optimize is really around bottleneck analysis for the pipeline and so what we've done here is you know we're instrumenting the pipeline almost kind of tracing it as if it were an application and providing some data around how are the actual pipelines themselves performing so what are the stages in that pipeline that are failing what is the duration of all of the pipelines the average duration in seconds and we're cutting that data by the individual stages in the pipeline so this is a really great dashboard to hop into and get a really quick understanding of you know what you know how are my unit tests performing you know how are they contributing to the overall pipeline time how are my integration test performance tests analytics tests you know some of those dashboards um you know that mike showed earlier and really understanding how they sit alongside the other parts of the pipeline and with that i just wanted to say you know thank you for joining today we have a q a right after the chat and most importantly i just wanted to thank mike for joining today and putting in the time to help show us all the great work that you've done in sumo yeah thank you for the opportunity to speak we're really excited with being able to work with sumo logic to collaborate together to form these dashboards and if you recall there was some history associated with this i spoke last year at the illuminate 2019 and showed these dashboards and we're excited to be moving into the software development observability dashboards great thanks again