 All right, everybody. Hello. Thank you for showing up. Let's just wait maybe a minute. Oh, no. Now it's noon. So I will start the presentation. So hello, everybody. Thank you for coming today. I'm Monica Wachee and I'll be your, like, facilitator, I guess, or your presenter, your lecturer. Let's hope this is not a lecture, right? That's so boring. If you can figure out how to use a chat, chat with me, I think I can figure out how to read it. And then we can actually have something more interesting, which is then a lecture. Oh my gosh. So today's topic, I called it SAS and Snowflake for cloud data storage, because I'm kind of in love with Snowflake. And I probably shouldn't be in love with Snowflake because what is Snowflake? They're just one company that offers cloud data storage. And why am I in love with them versus, like, Microsoft Azure or other cloud storage? Well, because they started flirting with me early. So you'll see Snowflake realized that SAS was going to have a problem. And they immediately started budding with them. And they've been doing that. I kept record of it since 2020. Okay. So that's why I'm, like, kind of in love with Snowflake. But I probably shouldn't be in love with Snowflake. Like I said, I'm just know them and I'm attached to them because they were cultivating a relationship with SAS. And I have a relationship with SAS. Love a relationship, right? But anyway, I do want to tell you that this is just mainly about cloud data storage. Try to move SAS to the cloud. And it doesn't necessarily have to be Snowflake. But the reason why I pick Snowflake is I know that they've been trying to work with SAS for a long time. So if you're going to do it, probably the most information out there is about specifically moving SAS data to Snowflake. So let's get started here. I just talked too much, probably. All right. So like I was just sort of emphasizing, today's lecture is really about, or boring lecture, you know, again, say stuff in the chat. I'll look over there. It's really about what happens if you have to move your data to the cloud. And when I mean your data, I mean your SAS server data, like you have a SAS server. If you use PCSAS and your data are not stored on a SAS server, then what I'm telling you about today is you probably don't have that problem. And Alex, I'll make sure you understand what the problem is. But here's the problem. If you have a SAS server and it's full, that's basically what the problem looks like, okay? So SAS made a deal with Snowflake early on as a cloud storage partner. You can use other environments. I'll get into that. SAS access, I did a lecture on that. When SAS access is a component of SAS that you can get in PCSAS or service SAS, that is an API, an application programming interface or API that allows you to connect, make an ODBC, which is an open database connection in ODBC. So remember, SAS doesn't like to play with other software, but SAS access has been around for a while. So you can open up this ODBC connection to a SQL server and pull some data in. Okay. Let's say you're not doing that. Let's say you're sitting at the SAS server. That's the server. You've got a big SAS server and it's full. Okay. What do you do? So then you can use SAS access to move data in and out of SAS, but that's not going to solve your problem. So you've got this huge headache of what are you going to do? So the question is, like, okay, my SAS server is full. What is my new solution? Like, where am I going to put all this? So other people are going to access it because I need it. And then also, how am I going to get there? You become a big problem. Before I go on, let me remind you that I have remind you if you haven't heard, but it's new if you haven't heard, is I'm holding a free online workshop. And it's the title of the course. There's an online course that I've already made. And the title of that course is application basis. And that course teaches people who are data scientists that didn't come through a business discipline, you know, like probably a lot of us on here who have MPHs or have master's degrees from like, or PhDs from like a health discipline, and you didn't come through a business discipline. Well, then you probably weren't taught a lot about computer applications generically, like how they're built, how they run. And if you want to connect SAS to any of them, you really have to know this. And this is becoming the bigger problem, because people collect data on their apps and what are app applications. So this is this course, the online course about application basics, I'm going to teach it as a workshop. You see, there's three sessions, Monday, August 7, Wednesday, August 9, and Friday, August 11. Each session will be two to three hours, because I don't know how many people show up. And so, and it's on Zoom, like you're doing right now. And what happens is if you, there's a link, if you go to the original event, or I'll post it on where I posted this post, you can sign up to go to the free workshop. And what do you get? You get to show up on Zoom, and I'll teach you this online course, you get the online course for free. And then after these three meetings, well, you sign up for a private wrap up session with me for 30 minutes. And that's your free workshop. So please sign up for it. I'd love to have some time with you, because there's space in the workshop for you to actually talk about stuff, like if you've got a problem like this problem, right? And then we can actually get into real problems you actually have, right? Okay, so let's get into the topic here. So why would you even, why are we even talking about Snowflake, right? And the reason why we're talking about Snowflake is because Snowflake is cloud storage. So I just want to make sure everybody kind of has a general idea of what cloud storage is like. So imagine you owned a server yourself, like if you've ever bought a server or you run a server, you know, it's just kind of like this boxy thing. And why is it called a server? Well, because it serves, it's really optimized for IO, you know, oh, you need this file here, blah, blah, blah. Okay. If you take a bunch of those, and you make kind of a farm out of them, you can stripe the data over them, right? So it's not like you're one little server that's serving everybody. It's like, now everything is sort of spread out. So it's easier to serve people. Now imagine you have a bunch of server farms in different geographic locations, because we have a really fast internet. Now we can serve you locally, we can do everything, we can save millions of acres, we can steal overs, whatever. And so you're like, wow, that's what cloud is, it really is a cloud, it's just big, and it'll just keep big, and server farms or whatever. Those of you into climate change, you're probably having a heart attack about how much fossil fuel that's using, that's a different discussion. That's actually a problem, like AI is having this problem or whatever. But let's just forget about that and talk about Snowflake So snowflake cloud server is kind of set up like that. So if you actually theoretically somehow managed to move your data into snowflake and into the cloud, not just snowflake like Azure or anything, then you've solved your problem forever, right? Right. So old people who are here know that you never solve your data problem forever. You just solve it until it's a problem again, right? And so I don't know what will happen. I can't see Microsoft going out of business, but Xerox went out of business. So who knows, like if you move it to Microsoft Azure, I really hope snowflake doesn't go out of business because I'm in love with them, like I said. So who knows what will really happen if you move your data to the cloud? Like what happens if it rains from the cloud or I'm joking. But really, I don't know the future. Like what I did at the US Army when I worked at the US Army as a civilian, I ran a data lake, right? And our server got full. And I said, can we get rid of some of this crap? You know, like we had some data sets on there from before I started that we never used or anything. So I had this fireproof safe, that thing was heavy. That thing was so heavy. Good thing I was at the army, I had plenty of strong people around me, but that was a heavy safe. But what I did was I actually wrote to disk the data from my server, which was like that boxy thing I was talking about. And I put it in that fireproof safe. And I was like, that's what we're gonna do. Okay, you know, like when you have a small apartment and you're like, okay, you know, we have a new couch, we'll throw out the old couch. You know, that's what I was like, okay, this is what we're gonna do. We're gonna just throw out data we're not using, old data, blah, blah. And the reason, like that sounds really old passion, but the reason I like to think about it is like US Army, like I don't wanna have data on these people I don't need to have, especially I don't want it on my server, because if it's on my server, somebody can take it off my server, right? So maybe we should care about the privacy of these people, but that's no fun. Let's just move it all to the cloud, right? So if you're the that's no fun type, you might wanna look at this white paper. So SAS Global Forum 2020, isn't that long ago? That's like in 20, that's like three years ago, okay? So that's, but if you actually pay attention, you'll find that SAS has not been cooperating or maybe not cooperating, but partnering with cloud providers until more recently. And so Snowflake got in early. And so this person, Jeff Bailey, an insiders guide to SAS Access Interface to Snowflake, he wrote this paper. Now, I'm just gonna tell you if you try, and he would tell you this too, if you try to follow this white paper today, it probably isn't gonna work for you if you try to follow it step by step, because everything has evolved. However, the good news is that, like you could maybe look at it for generically, what you need to do, and then you could get Snowflake to help you, okay? Because Snowflake, first of all, they're just like such a cool company. They have so all these people, they're really good at it and that they'll help you. But I'm saying they'll help you. At some point, you have to design your own solutions. They'll help you implement them, but you have to design your own solutions. That's one of the messages I try to tell you. Okay, which is why I made this diagram, okay? So what's the problem with cloud storage? The main challenge with moving SAS server data to cloud storage isn't the cloud storage. So I made this diagram to kind of explain the problem. So the SAS server is usually kind of like a junk drawer. Sometimes like in the olden days, we used to have phones hooked up to the wall and then you'd have a junk drawer next to the phone. So if you, it had pens in it, it had paper, it had paper clips, old keys, new keys, padlocks, ballpoint pen, all this crap in it. Because if the phone rang and you wanted to write something down, you'd just go in the store and you'd find something. Or maybe you needed to open a can. You'd go in the store and find something. Well, the problem with a junk drawer is it would get full and it would get full of crap. Like you'd have old Taco Bell sauce and there's stuff that doesn't really belong there. Like take out stuff or whatever. So you'd have to dump the drawer out and clean it out and just kind of throw some of the stuff out. However, if your junk drawer is a SAS server, can't really just dump it out and get rid of stuff because people are really into the junk on the SAS server. Now, when I've dealt with people, back in the olden days, you'd have a lot of SAS servers like in the early 2000s, I'd be like, how'd you get that to run so fast? They'll be like, I'm on a server and I'm not rich, I have PC SAS. So anyway, but they would be like, I'd be like, why are you storing all those years of data on there? Or they'd be like, give me your whole database, give me all the tables. And I'm like, no, I'm not giving me all the tables. And I'm like, do you have room for all the tables? I'd be like, oh, I'll just throw it on the SAS server. So I started getting really curious because I never ran a SAS server. Even at the Army, I wasn't running a SAS server. I was running a file server that had SAS data sets on it that we were reading in PC SAS. So I'd never run a SAS server, but those things are lightning fast. And so I'd be like, I remember when I was at the University of South Florida, I was at an office running PC SAS and I ran a proc freak with a Fisher that was like age groups against education groups. It was like three by four and it wouldn't converge. I went down from my office, which was off campus. I went to the server or to the computer room and I just ran, right? So that's how fast these SAS servers were. And so unlike SQL servers, they wouldn't really fill up. And so you, because they weren't connected, they weren't a bunch of indexes. So people just kept putting data on SAS servers and they were terrible about it. They wouldn't make a list of what it, people would leave their jobs and then your server would be full. And you'd be like, how do we archive stuff? Like it was just not really planned very well, okay? So anybody who has a SAS server right now probably has a lot of junk on it. So the first thing I wanna tell you is now is a good time to clean house. Inventory, everything on that darn server, I mean data, I don't mean macros or anything. Well, that's another problem, but just all the data sets you have, make an inventory, open up a spreadsheet, put attributes about when was the last used or whatever, see if you can put it in the fireproof safe because that'll already make that rock on that ladder be less big, okay? So that's the first thing. The second thing is then you've got all the data sets that you use. Now there are two kinds of data. There's like data at rest, like I would have at my data lake where it's just sitting there, and people would have access with once a while. But then some of you guys have some real runtime stuff going on. So it's like, you're gonna have to get it, that rock of data up that ladder into the snowflake, like through the internet into the snowflake, into that big cloud farm, that server farm in the sky. And then you're gonna have to access it, right? Like so in my data lake land, if it were me at the army, I'd be like, oh, okay, first throw out everything we don't want, then get it up to the move all each lake and move each data set up there. But then once it's up there, I'd have to be able to pull data from it like I used to. Well, if you're running a shop where you've got Viya and you're running like AI on the fly, like you're gonna have to do AI on the fly from snowflake. So there's just a lot of complexity. So I'm sorry, I had to perseverate on that diagram, but that's why I made that diagram. It's like, okay, so you've got all this data in the SAS server, it's like a rock, you're gonna have to carry it up like a ladder, which is kind of like what it feels like, cause you just have a little pipeline on the internet to get that data all the way through onto that big, into that snowflake environment or whichever one you're at. The bottom bullet here, there are a lot of handoffs along the way. I prepared this and I've been following Snowflake for a while. I just went and did a Google search today and I'm gonna show you one thing I found that's gonna show you, whoa, there is a lot of handoffs. There are a lot of handoffs, okay? And if you don't know what I'm talking about, I'll explain it better. Okay, so next, what the white paper says, okay, here's our first handoff. I'm not an expert at this stuff, but what that white paper says, and remember that's from 2020, it says to get SAS into Snowflake, SAS into Snowflake, you go through Amazon web services. Okay, so you go through Amazon web services, really? So I was like, well, wait a second, why are you going through Amazon web services? And then I started to realize kind of what's going on. Okay, so SAS is really not a database program. And so it doesn't have a lot of functions that database programs have. And one of the functions that database programs have, like sequels, is they can take the work of doing like a query and they can split it up. So like for instance, if you're just trying to like, if you've got like the BRSS data set and you just wanna pull the women out, which is gonna be about maybe half the data set, then if you're doing that in SQL, SQL goes, okay, why don't I partition this big table into these little like three pieces, pull the results from those three and put it back together. That's called threading. And you can, that SQL, that's why we love SQL, right? Okay, SAS doesn't really do that. And SAS has added things to make it better, but it doesn't really have those things. But Amazon web services, now let me explain to you my experience with Amazon web services. There's a lot more to it than this, but I'll just give you an example. I had a learner ask me to help them with an assignment and I live in Boston. I think they were going to Boston University or something and they said, well, we're doing machine learning and I'm like, I don't know much about that. And they're like, no, no, no, it's easy. I just need to use Amazon web services. And so basically what we did, they had a student account or something, we had some data in Excel, just the way you would have data for like logistic regression and SAS or something. And we upload it, you know, it's like you're basically running a regression model, like a much fancier regression models the way I see these machine learning ones. So we'd uploaded into AWS into that environment. And then we do click like this wizard and tell it what the dependent variable was and what the features are. That's what the independent variables are in AI land. And then we click something and we'd get results for all these canned AI models, right? Like they are canned. So we just look at like model results and stuff. And I'm like, that was really not that fun. But what Amazon web services does a lot more than that, they're the backend of a lot of, I believe a lot of transit, like online store transit, hence Amazon, like this is basically Amazon software that they're letting you use, which if you read Wired Magazine, you'd be like kind of suspicious of it, which is why I'm like going, let's say my SAS data is the army people back then. Do I really wanna pass that through Amazon servers? Come on. So I wasn't really sure if that's really what it meant. But what I realized is probably that's what's going on is the bulk loading capabilities. So bulk loading, what is that? It's basically being able to manipulate splitting up the work of doing the query, right? And making it so, well, you know how like, you can optimize yourself for bulk loading. And what I mean by that is like when you're moving, right? Like when you're moving apartments or whatever, you get a U-Haul and you bulk load the U-Haul, you know? It's not like you're bringing some furniture over to your friend's house, you don't need a U-Haul. No, no, you're bringing everything over to a new apartment, so you're bulk loading. And bulk loading is totally different than what you normally do. And so AWS allows you to set up like a bulk load from SaaS server into Snowflake, right? And so from the white paper, at the time I'm writing this February, 2020, the only object store supported for SaaS access interface to Snowflake bulk loading is AWS S3. Now I'm happy to say I think it's different now. I think we've got Azure. And so, you know, so this white paper suggests you're passing data through your SaaS server through AWS server, Snowflake really. So, and it looks like that. Now, if you're worried about like PHI and all that, well, I don't know what to tell you. I would have to look into that. And I thought reading this, it might not be ready for prime time, but of course that whole paper is a little bit older, right? So SaaS's tools have good IO in the SaaS environment. So that's why we all got very spoiled and just kept throwing stuff on the server, the SaaS server. It's like, it was hard to get in on the server once we got in there, oh, let's, you know, what if we need it, you know, just leave it there. We might need it. Well, you got to get rid of it if you're gonna move, right? So this AWS is basically to help IO. And what you can see here is this is from the white paper. And the white paper is kind of painful to read not because it's poorly written. So it's actually really helpfully written. And the person who did it, I just feel so sorry for them because they were really working with some new integration software, you know, like a new pipeline, but see this graph that shows the relative performance. So this is insert versus bulk load. And what's going across here, I've made, I made a bunch of these when I was at the army is like how big the data set is. And this is how long it takes to load, right? And this is the kind of stuff I was seeing at the army is like, okay, if we're loading a million records in PC SAS, okay, here's two million, here's four million, here's eight million, and now my SAS programmers are playing solitaire, right? And so this is kind of like what you get from both loading and so you can see, okay, great. Yeah, I mean, the more you load, it's gonna go up a little bit. It's nothing's perfect, but anyway, so and just like, you know, with data steps and everything, how you make your SAS code can change, you know, how fast this whole thing goes. And so basically, if you're here, I wanted to show you a few things. If you're thinking about it, my point is that you really wanna plan. So this is actually the white paper. So if you were gonna do this today, I would pull this white paper, but I would update myself. I would look at this white paper and then I'd Google everything it says to just see what has happened since then, okay? So what this white paper, I'm just gonna quickly go over it. It's, I think what it says is you start with your SAS server, you open up an ODBC connection using SAS access to AWS, like through two snowflake through AWS and I don't know how to do that, but I, and what they're talking about here is how a snowflake function support, like how snowflake supports SAS commands basically. And then, oh, and I guess snowflake has its own connectors, own ODBC, but this gives you, okay, then it says here, so this is how you connect to snowflake and then you insert data into it and then bulk load into it. You start explaining it here and that's where I started realizing, well, wait a second here, this is actually really hard. It's so cute here in the middle, he goes, just tell me what to do. And my heart goes out to him. That first part is just education, like explaining to the reader what you're talking about. And then the next one is just follow these steps and forget what I just told you because I know how hard this can be, bulk loading into anything. Just follow what I say here. But the problem is it's three years later, so you kind of have to not just follow what it says, you have to update it. So I was looking on Snowflake, here's Snowflake's website. So I thought, well, maybe I would see like a use case where they actually did a SAS to Snowflake and now they're using SAS Viya. So basically there's the data in SAS and then there's the analytics, right? So once you move your data to the cloud, it doesn't matter what cloud you use, you're gonna have to use Viya now. That's the whole point of Viya is basically so you can do SAS in the cloud. So it was weird because Viya was invented first before being able to do this. And I think it was because, like I never understand SAS as business plans, but I think it's because SAS wanted people to adopt SAS and they knew that they wouldn't really adopt SAS if they didn't make it so that you could do it on the web. Like, so for example, I don't know if you're familiar with Athena Health, but that's an online medical records platform. It's online, like it was built that way. I just read about it, I haven't used it. It's kind of popular that it was a startup, they're around here in Boston. Well, let's say that they wanted to, SAS wanted to sell analytics to Athena Health. I think they started in something like 2015. They were never not a cloud provider. So how would you do SAS? Like, I mean, you could give them PC SAS, but that would be bogged down immediately because they have a huge, I don't know how they do their analytics, but that's why I think they invented Viya first. And then they were like, oh crap, well now everybody's SAS serviceful, so we gotta get that in the cloud. And, but then at least you know if you move your SAS data to the cloud, Viya's already there. And you already pretty much know how to use Viya. Like it's a little awkward, but you know, it pretty much runs as always any other SAS does. So I wanted to look up to see, you know, just look what other people would say today, maybe not a SAS white paper, which are very extensive, but like, so I just Googled randomly and I found this pipes. So pipes looks like it's like Amazon web server. It's some sort of service that helps you. So I wanna just show you this diagram. Okay, so this is if you wanna move SAS data into snowflake and you wanna do it the pipes way, right? So see all these things down here? Like why is that coming over here? Right, I'm not sure. And then you have SAS coming over here and then that all ends up in snowflake. So I'm not really sure what this even means. So I kind of went down here and it says connect to snowflake, connect to SAS, use your credentials to all pipes access to the SAS API. So remember how I was just saying, do you really have to put your data in Amazon web services to get it to snowflake from SAS and the answer is no, you can put it in pipes, I guess. And so there's like, basically when you think about it, you're gonna have to probably need at least two things. This kind of an application to help you move your data from the SAS server and another application, the cloud storage application. And be careful picking the cloud storage application. Like if you choose pipes and you hate them for some reason, it's over. It's you migrated your data, it's done. You don't have to use them anymore. But if you pick snowflake and you don't like it, you're stuck because your data are in there. So that's part of why I like snowflake. See, remember how I told you I was in love with them? This is my blog. I looked back, long time ago, it's kind of an ugly post. I wasn't so good with the UX builder yet, but they had a data for breakfast. The breakfast was awesome. This is like, if you look at March 4th, 2020, this is literally the last thing I did before the pandemic. And then I never left my apartment. I'm not even kidding. And so because Genentech had some sort of conference, was it Genentech? And had a local spreading event. Thanks, guys. Good thing that they're so good with biology. But anyway, so I actually went to this. This was, I think the second one. I went to another one after this. Maybe it was a year later. I can't remember. Because one of them I went to, they went to demo it. And I just wanted to see, what does it look like when you use snowflake? And it kind of looks like Microsoft SQL in the cloud. Like, if you can imagine that. But it didn't really work at like, the first demo I went to, they just couldn't get it to work, right? And so I couldn't really tell. I didn't really write much about it because they couldn't even get it to work at the demo. But this is a long time ago, now it works. And then after that, so what I learned from snowflake, from this close intimate interaction is, I think they're full of money. Like, I still think they're full of money because that was just a great breakfast and they kept coming out there and messing around. Well, SaaS is also full of money too, but they don't know how to do this stuff. So it makes me think that if you go with snowflake, like if you're at a workplace that has a lot of money, like Google, and you have, or maybe not Google would have this, but we're like HCA, that's a hospital system that has a lot of money. Let's say that for some reason, you've got a SaaS server there and you have to move it into snowflake. Well, moving it into snowflake is not such a bad idea because these people are rich and they'll come and help you, right? I think that snowflake would come out, but they were really super nice to us. And so I really believe that if you sign up with them, now I don't know if other play, I don't know about the other pipe, what was it called, pipes? I don't know how what they're like. Yeah, maybe they'll really help. Well, maybe all these people are really helpful. I just don't know. But in any case, I just kind of fell in love with snowflake. They have such a cool logo and so hot outside. It's nice to get to snowflake. So I just have a few take home messages for you if you have to move, if your SaaS server is getting full. And that is that technology solutions only go so far. We might need more creative management solutions. So remember when I talk about inventorying, what's on your SaaS server? Like that's kind of weird to do on a SQL server because SQL servers are usually relational databases like you don't inventory that. But SaaS servers tend to be data lakes where you've got, okay, this clinical trial, okay, there's this pilot study, whatever. It could even be a file server, right? But the reason why it's not a file server is the SaaS server is because you wanna do analytics on it and it's too hard to load each time. My problem I had at the army, that's why I was going crazy. It's cause I didn't have a SaaS server, but I worked there from 2008 to 2011. I'm like, I am not buying the SaaS server now. I'll need to get rid of it right away, which is right away is like 10 years later, you know what I mean? So do we really need all these data anyway for what we're doing? Like I, do we need to move all the data into the cloud? And remember when stuff is in the cloud, it can get stolen from the cloud, okay? So if we're in SaaS and we're talking about health data, do we really need it all up there, you know? And okay, if we have to move it there, can we de-identify it first? Like is it possible to just keep maybe Excel crosswalks of IDs and stuff? But then the other thing you can ask is, can we somehow stay out in the cloud and preserve our SaaS environment? And if you're the kind that can do the shell game, you know, like I do in my apartment, well, you know, we got a new couch, you got to throw out the old couch because we're not getting a bigger apartment. If you have the discipline to do that, you're a good server manager. I think I vote for you. But what about today? Should a new SaaS shop just start in the cloud to begin with and be ultimately compatible with SaaS Viya? I would say, yeah. Like I would say do not set up a SaaS server today. Like I could be talked out of that. Like if somebody came to me with the right use case, maybe I'd be like, oh, okay, sure. But I'll tell you the main reason. And if you show up to my next presentation, you'll find out why. It's about migrating to R from SaaS. That's, you can't totally do it usually. But the reasons why, the use cases I'm gonna show you have to do with these people have a SaaS server, these places have a SaaS server and they got the bill. Every time you get another bill for the next year of your licenses and your SaaS server, that bill goes up. And the bill for R and Python never goes up. I mean, maybe you have to pay consultants, maybe you have to pay for equipment or whatever. Well, you still have to pay for that with SaaS. So it's like SaaS is getting painting itself in the corner. All right. Well, thank you very much. That's the end of anything I wanted to say. Like, you know, I forgot to look at the chat. Let me look at the chat here. So, oh, hello, Rudeau. I'm so glad you came. Now you got to see what I had to say. Oh, and then Mika showed up. I'm so happy she showed up. All right, so this is me. You can go to the LinkedIn event to download the slides if you want the slides. And again, I just, those of you who weren't here in the beginning, I just want to remind you that I've got this free online workshop you can take. Well, it'll be on Zoom. It's called Application Basics for SaaS Integration. But it's based on an online course that I've already made and you can go look at and you'll get for free if you come to this free workshop, which is about application basics. So what I realized is people who are trained outside of business schools or computer science disciplines, they may not really understand the design of applications. Like we all know SaaS pretty well. Maybe we know some other applications as well, but how are they made? And like who's making them? And how do they figure out what they're gonna do and all that? Well, the reason why you want to know that is if you want to do SaaS integration, just like I was describing how Snowflake is designed, like that's an application, AWS is an application. Well, it's really helpful to understand like the framework behind it. Like how are these things being made? So you'll have so much fun if you sign up for my workshop. We're gonna meet three times Monday, Wednesday and Friday, August 7th, 9th and 11th. Each meeting will be about two to three hours depending on how many people sign up. And then, and you'll get this, everything is free. You'll get the Zoom link, you'll get the online course and I'll teach it. Like it's not gonna be boring. You won't have to watch any of the videos in the course. You can if you want to, but I'll teach it to you. And hopefully we can have some really good discussions because what you're immediately gonna see is that all of this is kind of art, this integration stuff. Like whether you're talking about SaaS or any application, integration is where it is. We're analyzing data from applications. We're using SaaS to analyze Twitter data, analyze Facebook data. So I encourage you to come to this workshop because I think it'll add to your skills. So and with that, I will end what I was planning to present, but you can ask questions in the chat. Thank you for watching this video, which is part of the Public Health to Data Science rebrand program. If you are interested in joining the program, please sign up for a 30 minute Zoom interview using the link in the description.