AWS Supports You - Visualizing Your Cost and Usage Reports

[Music] hello i'm rob pagarita and i'm an enterprise support lead at aws and i'm based out of austin texas welcome to aws supports you where we share best practices and troubleshooting tips from aws support joining me today is aaron adel from aws sport can you give us a quick introduction eric yeah sure hi everyone i'm erin um i'm leading the go to market for something we call cost visualization basically you know trying to help customers get better insight and and value out of their costs and usage information and i apologize for my hair uh you know i tried to put some gel in it this morning but it's a little out of control i i just my general appearance i just apologize in advance to your viewers thanks aaron for the intro yeah today's topic was very near and dear to my heart too as an enterprise support lead and technical account manager uh we work with customers to help them reduce costs in aws and manage operation spend so that's we're going to be covering today is we're going to introduce you to the aws cost intelligence dashboard and kudos in order to visualize your aws cost and usage before we get into the details though a quick note to the attendees online please feel free to use the chat window on the right hand side of your screen to share your thoughts and ask your questions throughout the episode today i've got aaron myself here to try and help answer your questions on cost and usage so we really look forward to hearing from you erin do you want to go ahead and kick us off today yeah absolutely i'm i'm really excited about this and yeah definitely uh uh reach out with questions folks on the on the caller you know it's kind of fun to kind of answer some of these questions and make this a very sort of custom and bespoke uh demo um so yes as rob mentioned we're going to talk about the cost intelligence dashboard and the kudos dashboards collectively we call it the cloud intelligence dashboards um this is a little bit of a you know table of contents what we're going to do i'm going to talk to you about why do you need this why why why am i even here what's going on uh i'm going to introduce you to the cloud intelligence dashboard framework you know essentially what are these dashboards and how do they fit together and why and then i'm gonna talk about two of the kind of um spearheaded dashboards i would it's kind of a weird phrase but essentially the cost intelligence dashboard and the kudos dashboard are one of four dashboards that we have uh and i'll just kind of highlight them both and then i'm gonna do a demo of the cost intelligence dashboard so um customers customer use cases um you know i i get and rob i'm curious if you get these some of these questions too but i i have customers ask me questions like how do you drive a cost aware culture you know like yep what does that mean do you get that a lot yeah we do get some questions like that for sure about just you know really what does that mean and how do they do it so yeah i i'm not surprised i mean i think probably you know this is going to sound self-serving but you know the one of the first question is how do we lower our aws bill and it's you know the cost of where culture isn't necessarily about just having a lower bill i like to think of it as how do i track or extract more value you know essentially my bang for my buck how do i know that i'm getting really good value out of aws and i think you know you can break that question down into what's the impact of my savings plans and reserved instances um how do i visualize this stuff across multiple pairs i mean do you have you have a lot of customers with multiple pairs that seems to come up a lot i do have one that has multiple pairs so it's where this is really helpful yeah i think because i think you know you know you're an organization you have multiple payers you want to be able to sometimes see inside across all of it or you you acquire companies they have their own payer accounts there's a lot of different situations where i think that's relevant um you know they also want to see these opportunities to become more cost optimized sometimes they want to know what are the questions i need to ask of this data i don't even know where to start um and that's particularly relevant for customers who might be new to the cloud or uh you know might be new to how sort of cloud financial management works um what trends are important to track right this kind of speaks to how do i extract the value what am i looking for what's important to keep my eye on and i need to know when things are anomalous right did somebody just spin up a thousand instances without me knowing did they ask can they do that uh you know there's a lot of people who might not know what you know a lambda function is or what rds is or what gp2 is and there's there's these you know there's a lot of there's a lot of services in aws and you know sometimes you know folks who are in charge of the financials or in charge of you know making sure that everybody is kind of keeping track of how much they spend uh don't necessarily know what all of these things are so they but they still need that insight so how do you kind of give them that insight in the language they understand but then for the folks who do know what these things are the it managers the devop lead devops leads how do you give them that granular insight you know like cost per bucket what's my cost per bucket that's you know it's it's sort of you know i think an obvious question but it can be tricky to get uh that data so you know the dashboards help with that um my lambda cost per hour you know how am i how is data transfer working i get do you get a lot of i get a lot of data transfer questions a lot of data transfer questions as well yeah once that one's i think it's like hard to wrap your head around you have you know bits and bytes going in and out of there and where you know um so you know how do you get all this in one place is kind of the the grander question and that's where the cloud intelligence dashboards you know i call it a framework comes into play and this is kind of getting a little bit into the weeds here but um what happens is that all of this data that you generate right all of this um usage and cost is actually accessible through something called the cost and usage report rob which i think you're going to cover in a bit um but the framework sits on top of that and we use aws services athena glue quick site and the set of pre-baked sql queries um which is very helpful for people like me who like my sequel is not i would say i'm mediocre somewhere between mediocre and moderate at writing sql queries uh i don't know if you're good at them i'm not i'm not i don't i don't know i should i should not be programming people should just shut my computer off when i start doing sql queries but basically if you need to get this data out of the curve you need sql so we give it to you and then there's dashboard templates that are built in quick site which is a you know for those who are not familiar a bi tool from amazon that you can you can actually spin up in the console and these dashboard templates we like i mentioned we've got four of them sort of sit on top of this stack so it's a completely native aws stack and at the end of the day you just get out of the box these templates with lots of visualizations and data which i will go over uh trusted advisor also there you see that's a separate service that we have trusted advisor um i won't go too much into it but it it's covers security checks and flags and checks your iem key rotation and stuff like that really important for support customers there's a separate data source for data there that we can pull as well to create a trusted advisor dedicated dashboard so it's all based on the curve rob tell me yeah tell us all about the curve yeah so your cost of user reports right is alluding to everything's kind of built off this that comes into the cloud intelligence dashboard and what your constant usage reports are is it's the most detailed location or collection of data about your aws spend um and so you can enable these costs and issue reports and you could just have it you know leverage within athena so you can do queries you can collect information by individual resource id so you can find this specific ec2 instance you know talked out to this specific region uh in detail so it collects all this information uh for you based on all of your aws spend so you could analyze it using a ton of athena queries if you wanted to go down that right route and just export say from athena to an excel spreadsheet or something to then manipulate on your own um but overall this data gets collected it's delivered multiple times a day into that s3 bucket that you target this towards uh and then you know that really is what allows you to build those custom queries and reports we'll provide some links and information on you know if you just want to get started with how to create a cost and usage report in your account uh we always recommend when you're setting this up capture the resource ids when you set that up because that's going to be really key to getting that really detailed granular information out of your queries and such uh but we'll drop some links here about that and i'll turn it back over to you to kind of go over some more information on how we can build off cost and usage yeah thanks the the what's really cool about the cur is that it's a it's basically a big data dump it's a it's a it's now i wouldn't i wouldn't open it in excel uh necessarily yeah yeah right exactly so so um but that's that's a good thing in this case because it's it is a lot of data and you know i think when you get to a certain point with data you have to figure out okay i can't as a human go through this and do pivot tables and things how do i extract value out of this and what are some of the hidden gems and secrets inside here um and that's what i really like about this dashboard process is that you know on the on the all you know in the hole the dashboard is very valuable but even just accessing and extracting this data aside from visualizing it can be really valuable but i like to demo the cost intelligence dashboard and i like to talk about it first so as i mentioned before with the cloud intelligence dashboards we've got four uh publicly available dashboards the cost intelligence dashboard kudos uh the tao the trusted advisor organizational view and the trends dashboard i like to talk about the cost intelligence dashboard first because it was it was the first dashboard uh that that was created um and also i think it kind of it sits at a really nice place in terms of people like me who are not extremely technical and don't didn't even know what uh you know lambda functions were until i got here kind of thing uh and but also allows people who really understand cloud financing and fin ops and everything to kind of get the high level information they need um it's it's really good i i really enjoy demoing it for customers there's really kind of six primary sections we're going to do a demo of it so i'm not going to go deep into this but um it's got a lot of really good high level information what i do want to go deep into though before we demo it is the kudos dashboard so this is kind of like you know if you took if you took the cost intelligence dashboard buried it in the ground right you got some dirt on there and you put some like fertilizer on it got some water some sun and it just explodes into this bean stock uh and it it kind of grows it is it is an exploded view it is a very very very detailed very granular very in-depth dashboard um as as true with all these dashboards it's all still in plain language yes you do need to know what you know rds is and you need to understand you know why you're looking at your ebs snapshots but if you understand those you know why you need that granular information it's really easy to use it's organized by service so and kind of like service family so you can think about your databases in one section in your storage in another section it's pretty secure i mean this runs in your account um it's quicksite which supports your iam roles and access management there's no agents running anywhere it's just your data and your organization in your account and it's all aws native services so nothing leaves your account like i mentioned it's pretty in depth there's machine learning driven insights in there this is a really fun fact too i think is that there's you know you have the data that you've got at rest but then you can you can analyze that data and draw some conclusions or at least make some predictions going forward we were really just scratching the service on that too which is pretty cool place to start it's totally cost efficient all these dashboards uh that really the the the cost is just a quick site license which is 18 a month so for most customers including the sort of incremental serverless stuff that happens behind the scenes like the when athena runs and glue runs and everything um it's still like fifty hundred dollars a month it's a rounding error for a lot of customers it really is very inexpensive to run and you know it's a it's you can shut it down if you don't like it it's really easy to just stop using as well no you shouldn't though right but you can if you want to um yeah so here's just like some screenshots that i'm gonna have to squint to even look at i'm not going to go through them but uh you know just to kind of give you a sense of of the the the tabs in the dashboard that you get with the kudos dashboard and the sort of depth that you can go into into some of this data now rob you're probably wondering okay i don't even need to see a demo i want i want this spun up how do i do it right did i read your mind um the good news is that it it's a publicly accessible workshop today soon it's going to be a well-architected lab so you know keep your eye on that either way we spell out the steps to self-service this i mean literally people like me again who can just just use some of these surfaces for the first time like glue i've never even heard of glue what is glue i started using it by the way glue is awesome as an etl service just as aside i learned that you can just put pretty much anything into an s3 bucket and run a glue crawler on it and just create tables with the schema and access it with sql that's that's this is pretty pretty cool but but you know if you don't know what that is you don't know what athena is you've never used these services we literally walk you through like click here click there copy and paste this bit over there um there's also some automation scripts so it can be a pretty quick deployment i've done it with a lot of customers and it can go as fast as 25 minutes and you've got two of these dashboards three of them if you've got tau and four if you've got trends up and running and now it's just okay let's get everybody access to these dashboards and start getting some insight um all right so who wants to see a demo rob rob yeah let's do the demo it sounds great yeah i also want to add just real quickly on the kudos workshop uh if you're an enterprise support customer too like definitely reach out to your technical account manager because they can have a conversation with you about like helping to you know run through that workshop and get you going there with it and even though it's really easy to deploy uh we could also help you work through that so yeah that's that's a really key point yes talk to your tams they know this really well in fact um tam's built the kudos dashboard so and i should point out too before i demo these dashboards were built by some of the smartest people i've ever met and i've met a lot of smart people uh these people know this stuff they know the cost and usage data they have it memorized they understand customers they know what they're thinking they've been asked these questions a million times um i think what really i mean i'm standing on the shoulder of giants with these customers can trust the accuracy of these dashboards i mean they should you know you should always cross-check everything for yourself but um you know these have been there's been a lot of thought put into how these visuals get made and why they're there so i just wanted to add that to kind of bolster you know the support of it but let's uh let's kick off uh the demo here i'm gonna start showing you the dashboard so this is the cost intelligence dashboard here uh that i'm demoing again i wanted to demo this one first just because it obviously it's gonna be a little bit faster than demoing kudos which we could spend all day on but um actually you know you should check out the kudos dashboard we have a link here that's going to take you to a demo version of this dashboard that you're seeing and also the kudos dashboard the tao dashboard and the trends dashboard you can actually play with them you can you know it's all fake demo data so you can just go in and see how the visualizations are put together but when you spin up this dashboard um you're in this billing summary tab i'm highlighting here the forecast because this this looks a little strange right now but normally when you're on your real data this is going to be real forecasted amount but this is demo data which is uh you know we didn't have enough data to generate a forecast but this here this section here this is your invoice spend so this is like what's your bill right you get at the end of the month your aws bill and what's really kind of neat about this is that you can you can break it down so right now it's broken down by account uh which is useful but you can break it down by service you can break it down by charge type so you can see what was usage what was credits what was uh refunds etc uh savings plans usage ri etc um you can break it down by tag you could break it down by cost center you could break it down by business unit um there's a there's a lot and this is kind of the point of these dashboards is that it's really easy to slice and dice right if by different ways um by grouping in different ways the amortized spend graph here and all these numbers here this is more representative of what it actually costs to run in this case your your per account or your per tag or your per application your per cost center this is the actual cost of running it because you know the most common use case is you've got 10 linked accounts in a payer account right what happens if you buy you know an all upfront ri in your payer account and you've got 10 linked accounts that use some portion of that ri over the course of the term of the ri how do you charge back right this is the kind of thing how do you charge back to those linked accounts based on their usage this amortized spend can do that for you so you really do get the sense of the true cost of running these applications these these linked accounts these tags uh however you slice and dice it so if we go to the cost summary tab here um this whole thing is about your usage right so this is not going to include stuff that's not related to your usage of aws so for example your top spending account that would normally be your payer account but that's not very helpful information right because you know you might not even be doing anything in your payer account this is actually going to show your top account by spend in terms of usage these two visuals here which i really like show you what's moving what's changing right over the course the last two months you know am i using ec2 more am i using some service more uh it'll show you the top three that up changing services by cost and also by the difference in cost so you really do get a sense of what's shifting across again it could be across multiple payers um i like these numbers here this is the number of services you're using on aws it sounds kind of self-serving but the higher the number the better right because if it's stuck at like 12 and it's been 12 for a year you're not maybe taking advantage of some of the newer services that are out there and maybe not being cost optimized right i mean you're gonna have to you're gonna have to decide that for yourself but it's a good thing to keep an eye on this graph i like because it shows your usage which is this red line right here this is your overall usage of aws you know your gigabytes stored your athena queries your compute hours it's all combined into one number a number on its own which doesn't mean much but as a trend over time you can see is my overall usage of aws growing right or is it shrinking and then your cost so this is not causation but this is correlation you can see your cost behind this number and see how that is kind of associated with your usage um again these are all this is all about a story this is all about kind of scratching away at the service of this information as you kind of go through these tabs now what happens when something happens that you don't necessarily know about right an anomaly this little machine learning widget here is an anomaly detection widget that's built into quicksite and it'll surface anomalies and it'll make it really clear you know we think this is higher or lower than it should be based on your previous information you can dive in here and explore it on a graph and see which account it comes from and you know what service it comes from and really track it down or you can use this graph too i like this is just a daily cost graph day over day cost really great for spikes right like hey one day on tuesday for some reason we had this huge data transfer spike or i can switch it to by account and say hey account you know f had this like really large spike for a certain day and and i want to go track down and figure out why that was right i don't see any spikes here so i might just kind of move on from this point but uh it's really easy to drag just about any field in there and get a good sense of uh how you can slice and dice this further so this visual here right this is my savings right now what was it what does that mean so now we're getting into kind of apis and metrics and what am i tracking why am i tracking things right so here's a question this is a question for you rob do your customers think about only their ec2 coverage or with with savings plans in our eyes compared to on-demand or do they kind of go beyond it are they just looking at the coverage percentage now they look at coverage and they go depth into utilization and stuff too right because they want to understand overall how it's really being applied which ones are taking the most advantage of it some of them get really granular on the actual um level they really want to look at to see that information here so yeah and do they do they set like do they have a goal in mind is it like i just want to be 80 covered or something like that yeah on average that's what they usually aim for you know it always depends on what they're comfortable with right but um i've seen a lot of them that so far for myself that have uh wanted that nice 80 percent kind of coverage areas they still have some flexibility to use other things yeah yeah i mean that it makes sense because you know i think from the customer's perspective well this is kind of how you this is where the expertise comes in right because here's here's a here's here's the here's the the question i get asked sometimes is i i'm i'm 80 covered but i don't know what that means and also should i be 100 covered you know am i doing good enough how am i doing right how do i compare what i like about this this particular visual here the savings is that at the end of the day you are saving right so you purchased our eyes and sps and your coverage is what it is but are you still saving money and you know almost always the answer is yes and you can see that here in this graph um but you know i i i'm wondering you know i kind of i've been kind of launching into this now for a few minutes here and i haven't taken a breath to just check to see if people have questions at this point yet which i promised i would and i totally failed at doing so has anybody asked any questions so we do have one question from uh ray underscore db222 uh just curious odb222 i i knew it would be him just a question deploying the dashboards right is there cloud formation that's involved in deploying the dashboard uh probably some information with the kudos workshop as well but yeah pick your brain on that yeah definitely so there's there's lots of different ways you can deploy these dashboards now it used to be you just have to do it manually meaning you go and glue set up the crawler you go in athena set up the views you go on a quick site and set up data sets etc but um if you follow the workshop we have several options now that that you can use to deploy it including the cloud formation template so there's cloud formation that happens um even in the manual steps now with with with uh the way that glue gets set up because when you create a curve there is actually a yaml file that gets populated with your current you can use that to set up glue to crawl it and extract it so every day we work really hard to make this easier and easier to install so keep an eye on it because there might be some more exciting things coming in the future in terms of speeding it up um but uh but that's a really good question and and we'll definitely want to make sure people you know don't feel too intimidated about setting us up because it really isn't super hard it's it's it's i think the value of it just far outweighs the 20 or 30 minutes it takes to set it up um so anyway uh the the the the savings right the savings like like how am i saving am i doing and again i you know this is about tracking kpis and metrics and you know doing more than just i just want a lower bill right so your coverage you know you can track it here and and what's really cool is forget the absolute percentage per month but if you look at how it's trending over time you know like for example like last week it's you know i can't really see it i have to squint here my my screen is pixelated a bit but uh you know my my orange section here and this graph here represents my on-demand and that's obviously still a large portion of my overall coverage but i can see it's going down so i'm trending in the right direction other things you can do to control your your value is stuff like your previous generation instance then right i'm spending this little blue section here represents a little bit i'm spending on some previous generation instances i might want to think about upgrading those instances to get to get uh well to get what right so what's the impact of making these decisions what's the impact of buying our eyes and sps and whatnot that's what this graph is for i really like this graph because this is your unit cost right so this is what it costs on average per hour this line here to run ec2 now you can see it's it was at a peak at uh at that point and it's i've gotten it down right i've got it down here to some number i can't read because it's too pixelated um but it's it's going to be some cents per hour on average of running compute now that line goes down the less you use on demand the more savings plans you have the more ris you have the more uh current generation instances you have these kinds of optimization steps that you can take you can track how they impact you with this with this uh with this unit metric here so these things here affect that and that's what i like about this tab is that you can kind of see it all in one here you know um and you can compare it to your usage this line here is your overall usage of ec2 you know you don't want to get confusing you don't want to confuse it with with just a spike in usage or a dip in usage right because you might one month use the ec2 a lot less but you might have that usage that you did do might be more expensive usage right so it's not always you can't just look at the usage you have to you do the unit metric the unit per hour cost of compute is really important and then you can compare it to stuff like your customers what does it cost to have a customer what's your cost you know your cost of customer acquisition let's say you have subscribers how you know what's my ec2 cost per subscriber these are kinds of comparisons and things you can start to do with this data i like this elasticity graph because you can just see okay uh you know and this is all our this is all of our demo data so it's kind of flat and boring but you know i had a customer that looked this and said hey i noticed that we get we just have more going on on sundays i didn't realize that and it turned out to be true and they were able to plan forward and do more cost optimizing knowing that sundays was it was a big heavy use usage day for them so you start summary here this is along the same lines of your compute um you know i'll talk about s3 here on the on the left you also have vbs on the right um you know when you're storing objects in s3 you have different tiers right so you've got standard tier you've got glacier deep archive etc um this will show you how your objects are spread across your tiers in the previous month which is just useful to know but again what's my unit cost right how what's my average per gigabyte cost of storage and that's what this purple line here is so i want to get it down i you know i i could i don't necessarily think i can just store less which is this purple line here right this is my usage of s3 but uh you know if i'm if i'm you know using aws more and more and growing how do i keep that unit cost down so that's that's what that's what that graph really is about is um you know let's say you let's say you've started using intelligent tiering in october right and it's now uh you know march it's it's it's a few months later you can see the impact right of how it started to shake out what objects needed to get moved into into glacier and of course glacier per gigabyte is a lower cost of course there's different trade-offs but from a per unit cost gigabyte per hour average not per hour per gigabyte average cost of storage it'll go down your cost per bucket we talked about that this that's this graph right every time i show this to customers especially when they run it on their own data they're just surprised by something some bucket way more expensive than they expected and it's almost always a logging bug right so if you look at the names of your buckets here and you see logging i think you should think you know glacier um so you know that's a separate conversation but you really you can definitely get that kind of insight right off the bat with these um so reserved instance savings plans um this is your savings this is where your your savings are coming from right so you know across the ris and sps that you that you got which is seeing what savings you know i can this orange one is my ri it's saving me a certain amount and the blue one is my savings plan and of course you know this is demo data so we don't have very many savings plans in our eyes going on but as you can imagine you know that could you want to keep track of that especially as you start planning and strategizing for the future right you can see your savings per per savings plan or ri and compare you get to see where they go so in this graph here you can slice and dice it by um account by tag you know again where are my savings going right so you see where they're coming from you see how much you're saving and where they're going and then you get to look in the future with this graph and see when are they going to expire so this this savings plan is going to expire at some point in the future here in this date uh and that the that dollar amount that's printed there is what you're saving so you get an immediate impact uh information about what happens if i don't renew this your bill will literally go up by that amount uh so you get to not only see when these things are going to expire but by how much your bill will go up i feel like that's really really good information to know i mean rob what do you think right i mean is this is this useful and am i am i just easy no definitely useful information to have and i think you can give some additional insights into where that spend and when he's going right because we're typically like yeah yeah um i i definitely think that you know the more information we can give customers i mean they may not even know to ask these questions in some cases right i mean i i've now done this demo for enough customers to know that there's definitely a range and i just want to make sure everyone has this insight when they use aws because i think you know the the value here is really clear and also i think it helps you plan going forward to extract even more value so that's kind of what this graph is about now this looks awful right just a big weird orange square and and i and i do have to apologize because this this demo data made this graph look really uninteresting this is your average uh hourly spend that on on demand that is savings plan eligible right so um when you when you're considering okay how much like what's left right i've bought a bunch of our eyes at sps what what's where where else can i save this graph really helps so normally um when you're with this kind of graph what you're going to see is you'll see like a a more of a spiky situation but you're on demand may never go below a certain amount right so that you can draw an imaginary line uh somewhere on this graph right where you reach these these floors and let's say it just never goes below 30 an hour you never don't spend that right you're always at least spending an average of 30 an hour now some days in some uh hours you might spend way more right you might have these peaks of on demand but if you never go below a certain amount hey that's a pretty good place to get a savings plan right because then you can just commit to whatever that amount is and cover everything under that line with savings plan now if this this graph will be a real graph you would definitely want a savings plan because you're never ever changing right you're flat at 20 cents per hour on average or whatever this is and it doesn't change so excuse me um so i think that it's really important to keep an eye on this graph and see how your future opportunities shake out for more in additional savings so um this is a pivot table so there's a couple of pivot tables uh in this dashboard that i think really helps summarize this information make it accessible and easy for reporting so this is your reserved instance and savings plan tracking for your previous month so you have the the ris and sps on the left here you've got the type and the um the the date that they're going to expire and then all the way on the right here you've got the on-demand cost equivalent of those ris and sps you've got what you spent to get those rise sps and then you've got the difference which is your savings so you get in this one nice little graph the kind of summary that perhaps you're looking to give to your boss because you had to convince her to buy our eyes and sps in the first place and you can see see hey look this is how much we would have spent this is how much we did spend and this is the savings that we we saw um and then i'm going to skip the optics explorer for now because that launches into something deeper and i'll just talk about this mom pivot tab this is just a big giant usage and spend pivot table month over month so great for reporting as well any questions by the way that com have come in hey aaron yeah we do have one that came in from mr builder 10. so question is aws has a few cost-saving visualization tools and if they're using this cost intelligence framework and say trust advisor should both tools show the same number for overall uh cost savings and i if you want to take the water i can take so i think we have two different views that are going to be there right from trusted advisor it's going to give you cost optimization as far as what's existing in aws as far as like underutilized ec2 instances or ebs low utilization volumes and such right so a little different data than what's going to be presented in cost intelligence dashboard because it's actually based off your cost and usage data so what you're actually using not really making savings recommendations to you so much as giving you more granular visibility into how you are spending money um so you can kind of visualize that i don't know if you have anything more that you want to add as well aaron but i i mean that's the difference yeah that's a perfect answer i mean you know there there's lots of different ways to realize cost savings uh on aws and you know it's not always it's it's not always easy to see it or think about it in one place because it's kind of sometimes like comparing apples and oranges but um you're absolutely right it's it's um you know your your tau dashboard the trusted advisor dashboard like you said is going to show you all of your underutilized instances for example how much you're spending on them and this so essentially what you could save if you killed them the cost intelligence dashboard is going to show you you know your potential savings for uh adopting a new savings plan for example or at least give you a guideline of where you can start um i think you should use them all i you know i i really do i think that they're going to serve different purposes particularly in larger organizations you're going to have different folks who need to think about savings in different ways um you know people who are thinking about procurement uh and sort of what used to be more traditional capital expenditure versus people who are um you know trying to trim budgets or trying to find places to save or just modernize right and get their unit costs lower it just really depends on on who you are and what your role is in the organization and the good thing is we have tools for all for all those things um and and every day we're trying to make the experience better for customers and easier to see now that kind of begs it kind of it's kind of a good segue because um there might be some custom situations here and in fact this tool although everything i showed you is out of the box and when you look at the kudos and the tao dashboard it's all there's all these visuals out of the box right you don't have to do anything however you may want to do something right because you may have some custom use cases some questions you want to ask this data that we don't necessarily have a visual for that is great that's what we want you to do we want you to use this particularly the cost intelligence dashboard as a jumping off point this optics explorer tool is kind of the start of that so when you when you go here and you go into the controls you can use it like cost explorer right so let's say you want to think about what's my amazon rds spent i didn't see any visual for that so you can go in here and set some dates um and then it's it's going to give you your spend between those dates and then you just add a filter right so i'm going to filter everything out that's not amazon rds now i'm looking at my amazon rds spend bam right there's the answer but i kind of want to take it a step further right this is not the complete picture for me what about my storage portion of this i like the usage type filter it's a very granular field uh it's it's got a lot of stuff in it but if you just search for storage you're gonna start getting just the usage types related to storage and then i can group it by that and see okay this is my amazon rds spend grouped by the storage type and then i can you know figure out all right well this one's the one costing me the most or this one is you know this one's growing month over month what do i do about it yeah go ahead it looked like you're gonna jump into that yeah no i was just gonna say so we had a little bit of conversation going on with dragonite h3 in the chat here uh about you know like establishing access to the dashboards and stuff so i just want to give a little information on that too um but also just like you're mentioning like how we set this up originally is that cost like aware like engagement for our operation stuff that's really the important part that we can leverage with the nice thing that it's being in quick sites you can kind of grant access through a few different methods and stuff so you can give them access just by publishing that dashboard and giving them access through i am or if you use active directory already on premise you could actually give access to quick sites react directory so a lot of different ways you can get it that you can maybe get a more cost aware and cost conscious uh team it's always easier said than done sometimes but the area we want to just kind of touch on so yeah totally i i and that's a really great point um you know i think one of the one of the important things like you mentioned about the cost of work culture is creating a culture in the first place right and so you know it's it can be financial data it can be it can be um data that's not necessarily um you know for everyone but there's definitely going to be groups of people that need to see certain things and and quicksite you're right quickside has a lot of options for sharing these dashboards there you can create non-editable versions of these dashboards so let's say rob that you were in charge of accounts you know linked account one through ten and i was in charge of linked account 11 through 20. i wanted you know our boss she wants to give you this dashboard but only for your accounts and she wants to give me the dashboard only for mine it's you can totally do that you can literally just fork the dashboard and create the rob version and the errand version and each one is filtered just to show data from those 10 accounts for each of us um and there's all there's lots all sorts of different ways you can do that too you can create a mapping table that maps rob as the account owner for these accounts and me as the account owner for those accounts and you could you could filter we could have like competition who's saving the most right um and lots of stuff like that um so um you know i think what's really cool is that it's all customizable and and that which is kind of what i was getting to here so let's say you wanted to create your custom a custom visual right this is going to be a permanent version of what we just built in the optics explorer uh but permanent right so this this stays in the dashboard as a visual so um first i would just you know start with my usage date my x-axis is going to be dates great way to start a graph you know especially if you're new to it i'm going to do cost as my y-axis right so now i've got cost over time i like to change this to a monthly aggregation and make it a bar graph it's a little bit easier to see and now i'm just going to basically filter this by uh by the service right so i want to filter by amazon rds and then i want to filter the usage type so i'm going to do the same thing here i'm going to look for storage as a usage type and then just apply this filter and then what i've got going now is um you know that that same graph but but i want to group it right so i want to group it by usage type so i can just drag it up into this group by color by field here at the top and it's going to do exactly what i needed to do there's my groups of storage i can see it here but let's say one i want more right okay i've got that answer but maybe i want to know it by a tag again or by region uh anything and just drag it in there right and what's kind of cool is you can play with this you can search in these these fields uh for you know various types of ways to slice and dice this data and just drag it in and see it see see if what you're seeing is useful information um and then of course this kind of leads to all right well what about filtering or what about um you know doing multi-dimensional types of visualizations the possibilities are somewhat endless um and and again like we we really want to encourage customers to play with this data you can't break it you can't mess up your data it's not like you can you're not going to crawl back into your curve and start thinking stuff right uh that'd be really weird but what you can do is explore your data there's all sorts of cool stuff in the curve there's all sorts of cool ways to slice and dice this data i mean every time i do a demo of this for customers they're always asking well what about you know how could we show this kind of information you know i want to see um here's a common one i want to just see my my cost per upload to s3 bucket by my customers so you can just put a filter for put operations for s3 services you can show it over time and then you can group it by account you could group it by tag you could group it by anything right by region now this is where the kudos dashboard really comes in handy too because chances are the kudos dashboard has the visual you are looking for right these are the visuals you are looking for um star wars is very cool if you haven't seen it then that's your problem not mine so i'm just gonna put that right there that was a star wars reference uh and and what's really neat is that in kudos you can you can you can look you can hunt in there for examples of visualizations and say yes that's what i need i don't need to recreate it it's already there maybe it's also just partially what you need so you can click on it and see like how is it filtered how is it organized how did they do this and copy it so there's there's there's a whole world of exploration that we want to encourage customers to really kind of start to take ownership about over their costs and usage data um a a colleague of mine calling the term controllership you know how do you how do you how do you give the control and the ownership and the accountability to your teams right um and and and and how do you balance the folks who know what gp2 store gp2 is with the folks who are thinking about you know various financial costs amortized processes and these kinds of financial reporting mechanisms that that that the people who know what gp2 might not understand right there's a kind of gap and i think that that if you drew a venn diagram that cross-section i think is what we call fin-ops at this point it's like the merging of financial people and developers and this is the these dashboards are here to kind of open up this world to everyone and we're always improving we're always looking for feedback you know um we're always you know this is not this is by no means supposed to be the end-all be-all this is the start of a journey this is scratching the surface this is the tip of the iceberg this is the other metaphor insert your metaphor here for that for that sort of related thing um and uh and and you know i i i just hope customers you know give it a try takes 25 minutes um it costs very little and it's it's easy to walk it back if you decide you don't like it so did we did any other questions come in while we were chatting uh no other questions right now that came in yeah just to re-emphasize what you said yeah it is pretty easy to deploy i've set it up myself uh and it's really valuable just so helpful to be able to view um with all the different dimensions of stuff that's like you mentioned in the curve that's just kind of there for cost and usage information so that's awesome and do you think i mean we have we have a lot of tams who set this up on behalf of their customers so that when they meet with them they they have these insights um would you encourage customers then to ask their tams you know to to do that for them as well as help them set it up on their own accounts which do you think we should encourage customers to do i i like encourage customers to set it up within their account for when it's low cost and two then they can kind of get access to whoever they want to however they want to right within their account so they can delegate out who can access to what and kind of go through all the customization themselves so yeah that that the delegating of access thing i know i know that was asked before and he answered it um and i thought that was a really good kind of um it's i didn't you know it didn't cover it in the in the powerpoint or in the demo very much but um do you think that you know when it comes to cloud financial management and accountability how should customers think about that access delegation and you know where should they draw the line or should they should they be more sharing or or or clear with who should have access to this information um you know i think it depends on the organization overall but i like to encourage my customers to really make sure that everyone can be aware of what things cost uh and be able to see how it impacts you know overall in the environment what overall spend is it just gives the ability then to continue to innovate to reduce costs in other places so they can put it back into where they really want to grow so giving that higher level of visibility across the organization just gives everyone a little bit more awareness about it so yeah that that's i think i think that's really key um but yeah that i mean you know that's it essentially i really do i you know i'm hoping customers get a chance to to check out some of those links um we can talk about more about the deployment too if you think that's worthwhile i can just kind of go through a little bit more about what's happening behind the scenes or uh or we can talk more about some of the visualizations what do you think would be most beneficial at this point uh yeah i don't know i so we had a little conversation about the deployment steps of it um if you had just a brief bit of information on the actual deployment we could probably cover that for just a couple minutes um today okay so in the link in the workshop link you will see you know when you go to it you're going to see a set of prerequisites the prerequisites are supremely important uh and and i think that i'll just highlight some some of the things you'll be doing so you're going to set up the curve the cost and usage report now if you've already got it enabled chances are you're going to have to create a new one because you need to set it up in a specific way it needs to be in amazon athena format slash parquet format uh you need to have resource ids and it should be in the hourly granular granularity um and it's okay if you don't if your car is not set up this way don't worry about it we can backfill it just put in a request with your tam uh we can put your your you know historical data back in there so that you're not starting from scratch um your glue crawler when you get that set up is gonna it's gonna run and scrape that curve and you're gonna set up athena basically where you're to copy and paste some sql queries here that goes five of them and uh they're gonna run against what's in s3 through the tables and schema you set up in glue once you've done that and and you know the automation scripts are going to do this for you so you know you might not need to worry about this uh you're going to go to quick site and create data sets and these data sets are going to then you can this is where you get to schedule like how often do you want this to refresh i recommend once a day every 24 hours um that's when you get the most eventually consistent data and most sort of up-to-date data without you know going overboard with with refreshing as well and then once you've got those data sets set up uh you can you can you have to post again the automation scripts will take care of this but you post a little json through your command line interface to quicksite to pull the template and basically create the dashboards uh and then then you're off off to the races but you know um i i don't think it's too hard we spell it out really clearly in the prereqs because those are the most important steps um and you know obviously we're here to help so if anybody gets you know runs into issues definitely reach out great yeah thanks eric um so everyone yeah today we looked at the cost intelligence framework and some of the different dashboards that are available within it how you can leverage those to visualize your costs and usage reports so just want to say thank you to our moderators for posting the registration page for our upcoming episodes which is aws supports you dot splashdot.com so make sure to bookmark that page and we hope you join us next time if there are any questions that were not answered today you can post your questions on forums.aws.amazon.com you can also email us any feedback to aws supports you at amazon.com really want to hear from you let us know what else you would like to see on this show i really also want to thank you for joining us aws supports you and happy cloud computing you

Channel: Amazon Web Services

Views: 1,145

Rating: 4.8571429 out of 5

Keywords: AWS, Amazon Web Services, Cloud, AWS Cloud, Cloud Computing, Amazon AWS

Id: Gq15yuPJ2Kc

Length: 52min 24sec (3144 seconds)

Published: Tue Jul 13 2021