Hello there, everybody, and welcome back to Redgate. Why do I sound so... I sound like I'm doing some kind of crazy intro affair. Roll up, roll up, ladies and gentlemen.
No, of course, it is... Right, you've only just noticed. Like you do this every time. It's true, and I do sound like...
I feel like my voice is trilling, just all that a little bit more. Anyway, welcome to the world of tomorrow. Well, now you sound like you're about to introduce, like, tales from the crypt. I think that was a future armour line.
Anyway, that, of course, today is Redgate, DP. Elle! Ooh! That, of course, is Redgate 33, so we are currently at half of Chris Coswell's age.
That's right, I can make age jokes, too. Hey! Three times yours. Oh!
No, that's... That's going to infringe some copyright that's going to come down on our heads. Oh, welcome to DBL, folks! I do.
So do it again. Yeah, exactly. How many cartoon characters can we do the news as? Raggy, cataloging!
LAUGHTER That's how to get all interviews, by the way. Just make sure you're playing just a beater in the background. That's a fair shout, actually. There you go, top tips.
Anyway, don't say we're never useful. Welcome back, of course, to Redgate, DP. Oh, yeah, but sorry, for all the people that was useful, not information for, it's not useful for us. It's useful for, like, the less the one beside the world.
That does do do do do do. Yeah, that's one of the interviews. That's fair. But, of course, that's right.
Welcome, everyone, to Redgate, DBL. It is a pleasure to have you here. Thank you for joining us. Of course, I am, as always, Chris.
Your welcome. It's a pleasure to be here, too. My name is also Chris. No, see, they go proper Chris and Chris.
So that's fair enough. I like that. Now, Chris, today we're going to be covering a topic very close to my heart. And I think every topic close to your heart.
You're close to my heart, Chris. That didn't answer the question, but I'll allow it. It's good. You sound like you did catchphrase.
It's good, but it's not the one. No, of course, today is a very, very special day, of course, because we are celebrating. On birthday. No.
Wait, it's not your birthday yet. That's a true one. Give me a second. Before, anyway, no, of course we are celebrating.
And that is because Monday, 21st of June 2021 is a very, very important day in the calendar. And why? It's International Giraffe Day. Yay, International Giraffe Day.
Absolutely. So do be sure to go out and learn some facts about Giraffes. If I was better prepared, I would have prepared some facts about Giraffes myself. Unfortunately, I haven't.
I'm looking at Chris Kay and hoping that he's maybe prepared some facts about Giraffes. Did you know that there are 68,000 Giraffes left in the world in the wild in 2020? Oh, wow. You declined 40% in the last 30 years.
So there you have it, folks. Giraffes, incredibly important parts of our ecosystem. Please do feel free to go and find any giraffe charity of your preference. Sponsor a giraffe, name them Chris, whatever you like.
Go away and... I like giraffes. Don't get me wrong. I don't want to see giraffes go.
Are they an important part of the ecosystem? A little leach pin that holds everything together. No, that's mosquitoes. But no, go away.
I saw a study recently about saying that mosquitoes being removed from the world would not in any way impact anything. Except to make this bump on my hand from where I got bitten last night, go away. Thank you, Mosquito, wherever you are. But of course, no, go away and feel free to sponsor a giraffe.
We love giraffes here at DBL for no other reason than they're pretty cool. But of course, today we're actually not going to be talking about... Well, we are going to be talking a little bit about giraffes, but we have correlated this with our actual session. So today we're going to be talking about cataloging your data and why it's important.
So we're going to cover things like what do we mean when we say cataloging your database specifically. And the way that we're going to approach this is of course with a giraffes view. We're going to get that really top level view and we'll talk about how the giraffe relates to classifying your data estate in due course. But just before we do that, Chris, it's time for beer.
What are you drinking today? Today I have a Greek line caviar. I don't know if you can see that. I can see that.
It is a maple and pecan Danish stout from a turning point. Wait, wait, wait, wait, wait, a maple and pecan Danish stout. You're just trying to outdo my blueberry muffin. Stout.
Maybe. Maybe. Okay, fantastic. Well, I today and drinking a St.
Ostele brewery proper job. That's not what you have to talk about. You can't actually see. There you go.
There you go. Powerfully hopped IPA. And the reason why I'm drinking this particular beer today is because... Because it's an alcoholic.
But of course, the reason why I'm drinking proper job is because a lot of people don't do a proper job when it comes to governance and finding where sensitive information exists across their estate. And I mean that in the more general sense, people do a half a job. They find the Excel sheets. They find the emails and everything else.
A whole bunch of structured and semi-structured data falls by the wayside, but no, they think they know where everything is. That was the biggest shoe horn. So if I tried to get my wedding shoes three years ago, Chris, that was great. Well, you shouldn't have ordered size five shoes, Chris.
So let's get these open. Ready? Three, two, one. Oh, that was terrible.
There's no chance you'll hear that. You're terrible. But I'm going to pour it. I'm going to pour it anyway.
Using the headset mic makes this interesting. It does mean I get to smell as I pour, which I thought would be more prominent. Well, this one actually, this proper job smells absolutely divine. It smells like a really fruity, really, really hoppy IPA.
Oh, it's 5.5%. So it's pretty solid. It has a really nice color to it. I don't know if you can see that.
If anyone's watching the recording or if anyone's currently drinking a proper job, it smells great. They've done a proper job. It all zoomed backgrounds and held objects. They work really, really well.
Barbara, job. Oh, that's really nice. It has a crisp, bitter finish. And I agree with this.
It is Chris approved. That's it. Well, this is really good. It smells not what I was expecting.
It smells like a sour and quite countryside-y, but tastes like a sweet stout. You've got yourself a sour stout. It tastes sour. It smells kind of.
It's sure it's still in date. Let's hope so. It doesn't go out. It just gets more lively.
Hotent. Oh, this is very nice. I could definitely drink a second of these. It's a good thing.
When we bought them, we actually bought them from a local supermarket that rhymes with Bessco. And thanks, Chris. Yes, exactly. And I had a choice of one 500 millilitre bottle or four big cans.
So we were kind of unenying. I decided for the one bottle. I'm glad I did because it's, as I've said in previous DBLs, it's quite quite co-operable, which means that I will get through it somewhat fast, somewhat rapidly. Compared to all the other types of beer in the world that you drink slowly.
Yeah. Okay. Fair enough. If I don't like it anyway, of doing things quickly.
Yeah. What a tenuous link. It's perfect. I love it.
So absolutely. So today we're going to be talking about database classification and cataloging your data. And it's, you know, we've, we've kind of alluded to it at times before. We kind of, we've spoken about it a bit, but we've never really spoken about it at length.
Like we never really talked about what database classification means and how it can be used. And the problem is when it comes to databases in particular, and we'll, you know, just as a bit of background, right? The number of times we get the questions, like DBAs or developers, IT architects, so data architects, largely enterprise architects. So we might start getting the questions like, Oh, where is our sensitive data or how much risk does this system pose?
Yeah, if we're moving this system to the cloud, you know, how much sensitive data are we storing on that particular database or on that particular instance. And the problem is when the something that we've observed and, you know, might not be true at all. It might just be a subset of people. Was it your mathematician?
Would it like set bias or something like that sample bias? Sure. There you go. So it might be a sampling bias, but a lot of the people that I come across have this approach, this almost giraffes I approach where they're way, way, way up bear with me.
They're way, way, way, way up on the weather giraffes eyes and they're looking down and they see grass. And they're like, yeah, that's all grass down there. That is 100% all grass. But the problem is what they don't see is the gaps.
Okay, they're seeing it from so far up that they think that simply by classifying your estate, as I simply, definitely not, you know, finding your data citizenship, I think, is one of the major kind of data classification tools calls it is not an easy thing to do because you've got so much to find everything from Excel spreadsheets through to, like I said, emails dot text files that one time you took a screenshot of a report output and then dumped it into an email and sent it to yourself at home. There's a billion ways that these things can go wrong. Well, not wrong, but how we can miss stuff. Right.
So you need to have the right way of doing it. But your rough side view, I'm still exactly over there. But the problem is a lot of people worry about the stuff that is easily missed, right? They worry about Excel spreadsheets and documents and xyz.
So that's where like 95% of people are putting their efforts, they're going, Oh my gosh, where are all these Excel files? Ah, you know, something like 1940s reporters. Oh my gosh, where are all these Excel files come from? Now you hear me, Shane.
But I was a very good Atlantic accent. Oh, thanks. Yeah, yeah, if you don't like my accent, you're going to be sleeping with the fishes by all anyway. Anyway, so the problem is there is suddenly if there's anyone in the world who actually talks like that, please get in touch with us.
We'd like to hear from you. We want you on the podcast. It's just like, Hey, how you doing Trevor? Yeah, I'm doing all right.
You see, it's a speech impediment. It's just what I got. Me and the boys. Just basically the Antilm off from wacky races.
Yeah, exactly. Boop, lagging whiskey. Welcome to DB whiskey where, you know, three grams and we're down. But yeah, I mean, 95% of people are not really paying attention to the thing that's staring them in the face.
When you think about an organization and how it gathers data, if only we had a place, I guess you could say a base of some kind of, you know, some kind of store of data that, you know, we just we keep and use. And use and serve our customers with a wait a second. There are these relational databases that we've got coming out the wazoo that we've just ignored. And you know, a lot of a lot of tools and solutions that scan for, you know, semi and unstructured data.
They don't necessarily have everything in place to tackle databases because there's almost this kind of implicit belief that people know where stuff is in a database. But the number of times I say to people, hey, how much do you know about this database? The answer is, yeah, not much, or, oh, I'll go ask Dave or Caroline Suzanne because they know this system and it's, you know, it is just another major silo of people's understanding of their data in an organization. Yeah, so I think that's an interesting point, like unclassified unstructured data classification is a really hard problem.
Yes, it's not easy. But that's not really what we're talking about so much. It is a tough problem and it's fundamentally different to a structured data classification problem. Yeah, we know where all the data is because there's such different approaches you use different technologies for it.
So anything that says, okay, well, you just use the one magic solution for the whole lot is just not true. Yeah. That'll be a Frankenstein's monster of lots of different things to together most most likely. What you need is good is a, they say that, you know, a lot of people, I think if we refer back to the state of database DevOps 2020 I think it was that had this particular statistic and it was like people had gone from trying to shoehorn everything into one, two, three tools to being open to actually taking on 10, 11, 12 different tools to achieve their goals, because, you know, you might have.
So good example when we talked about monitoring databases, you know, there are times where you will need something that monitors your databases, because it gives you insight into, you know, weights and deadlocks. It gives you insight into like 10 DB contention, all of these different things that you're not going to get from a network or an application monitoring solution. And the same is true about data classification. There are some amazing solutions out there for finding your semi structured and unstructured data so pick one off the just off the tree, Calibra or one trust, you know they've both got solutions for this.
There's obviously also Microsoft have purview as well, which does a pretty good job to but I've not I've not delved into it all too much. So I can't really speak with much authority there. The point is there are solutions for this. And then comes the database.
And the funny thing is, when we say, okay, understand the database, there's such a thing as finding where sensitive information exists. And tagging that information, but it's what we do with that information that makes the difference. And it's not enough. And then you know, governance can sit, it can sit all day long and take a look at their dashboard and say, ah, we've got 1,198 excel sheets that are that hold sensitive information and the same is true of databases.
But what's the process where those excel sheets new excel sheets that get created we capture. But what about database changes, what about when one of our columns gets removed or renamed or moved or you move data around to other tables or XYZ. And we'll get to that when we talk about what you do with classifications, but let's bring it right back. So Chris.
When we say classifying the database specifically. That's interesting because I'm thinking about this like with that old school records keeper hat on and you think of like an individual record and you think like that record has to be classified and identified and then. Yeah, maintained in the correct and proper way. So for example, a document you get a document or an email that comes in, you classify it and then you store it.
And back in the olden days before computers, that would be a letter that you put in a filing cabinet. Whereas these days I don't know why Bristol is my mic go to old. I was going to say that they know what computers are in Bristol. So you have your file you put in the filing cabinet and that filing cabinet is kept for 10 20 years and then it all gets moved into a separate warehouse and then 10 years after that it will get spin.
Fine. Okay. In the digital world, there's a slight difference, which is the document that comes in. It doesn't get put in the filing cabinet.
It's a it gets kind of tagged with this metadata and you have this concept that it's part of the classification. Not only that, but the record itself. So if you think about an email, that entire thing is one record. If you think about a database, what is the record is probably going to be the row, right?
Or is it an individual as a whole within that database? Like is it Chris on wind everywhere he appears? Is that the record? It's very hard to say.
So maybe you just need a completely different view of this. So what a lot of the technologies out there that will think about structure content. They'll still have that records management view of the world whereby you take everything on, you classify it, you keep it as a record, and that defines how that record is kept on the term with a basis that just doesn't work. You can't really take like a row in a database as a record and maintain it that way because you'll suddenly find that you've got 100 million records just from one database.
It's kind of nonsensical. So the interesting thing there is that it's actually easier than going is what the data community call our bar. A bar, a bar, row by agonizing row. So the way that we look at this is instead, we break it down into something even more generic, and there's something even more generic about rows in a database is how those rows are constructed, like what is the structure of those rows.
Now, if you think about it, if you transform it onto its other axis and you start thinking about columns, you know how a row is made up because when you look at the row, you know that the first entry in that record is going to be an ID. And the second is going to be a name, and the third is going to be an age, and that is defined by the structure of the table on which the record is held. So it makes sense that instead of saying, okay, well, each row by row will figure out which one sensitive, what we do is make a more generic approach and say, okay, anything in this column is going to contain a sensitive value. Now, obviously this breaks down when it comes to things like JSON, you know, if you're working with a no sequel database but we're talking relational databases for the moment.
Structured relational databases, exactly. And the interesting thing here is that, because you know how these rows are going to be structured because you know how the rows and columns fit together, it's pretty fair to say that if we've got a column called date of birth. But yeah, in that column in basically every row if it's not null, then almost every single row is going to be a sensitive value. So, now, what we mean when we say cataloging the database is actually specifically around tagging columns like we've just described it's easier to tag more generic columns than to try and go row by row.
But the problem is, you know, we still have columns across tables across databases across instances there's still a lot of columns to do. But it's still important to do this process because once you have all of the columns tagged, you then effectively go about saying okay when I look at this database when I get the question, what kind of information do I hold, you can look at a date of birth. If you're not a single column, you can look at a database in a single glance and say that database is full of personally identifiable information, approximately 65% of the columns held on the tables in this database are sensitive, because we have them marked down as x y and z. And that's why it's important I suppose to always define a good taxonomy.
Because if you don't have the right tags and you're just tagging these columns as sensitive non sensitive, and that doesn't really give you any more insight other than yes, someone theoretically. It is is a very subjective job. So if you go into it and you say yeah that's sensitive, and then you look at it, it might be a date signed up for a service. Well, maybe that's sensitive, maybe it's not, it could be a very indifferent level you could have multiple users who signed up on that date.
And that's going to be less sensitive than a date of birth. So what we know is interesting one there right context, it is like subjective but it's but your goal I think is like a, I don't know the business's goal is going to be to try and make it as non subjective as possible right they want to be able to apply rules that will. be able to say this is in this is out this is in this is out. Exactly but with those rules and with that approach still comes a level of where we need to apply a number of tags to ensure that the metadata associated with these columns is appropriate right it gives us insight.
And that's the thing when it comes to databases is not everyone has that insight. And if a DBA leaves are dead leaves who has worked with that database for a number of years. Yeah, then you lose a huge pool of knowledge. And that's a that's a terribly scary thing for a business like having any kind of knowledge.
Just stored in someone's head rather than an application or on a knowledge base somewhere. Yeah, terrifying. Exactly. So when we look at the giraffes of your dress I view from the top of the organization you've got this, you know, board of directors you've got this CIO looking down and thinking, hey, I can see all the grass.
So I know where the food is I know where the mud is I know where the water is I know where everything is. But what they can't see is the ants nest the fire ants nest right which is the database it's a mess nobody knows exactly what's going on in there nobody knows where everything is and then a DBA leaves and all of a sudden you lose your your expert in that particular set of databases and hand over documentation is always appreciated. But does hand over documentation often work out well for everyone. I can't think of any situation where I've ever been given hand over documentation where it's covered everything you can't substitute human knowledge.
So what we need is both the giraffes I view we need something that helps us manage our wider estate like things like, you know, per view, like, Calibra like one trust, but what we also need is visibility around the database. So we need something that allows us to ensure that we are keeping the metadata about these columns up to date. And realistically, when that fits into kind of organizational compliance, this is where things look very different because people start thinking, okay, you know, it's really easy to be compliant we just find all these semi unstructured data we've got, even the structured data. But then we start talking about differences and that's I mean you and I have both come across situations where people have put metadata about columns into an Excel spreadsheet, for instance.
Yeah, you know, it's, it's, it's, it's probably the, the, the first book all for a lot of people are kind of experimenting with this put it all in an Excel and I've seen some really good Excel documents. It's worth it. You know, do actually a lot of the work for you, but ultimately, you know, someone has to maintain that and you almost kind of transfer the knowledge from one thing to another. As soon as the person makes the Excel leaves, then you've got the problem.
Exactly, because they leave and you lose access to it or people forget it exists. People forget to update it. What's just how it works. Yeah, exactly.
So really the problem is, is that it becomes an additional administrative task. Oh, yeah. That's the killer. The moment you give developers the moment you give devs the moment you give compliance and the moment you bring compliance and governance in you know compliance and governance care about more high level specifics right they care about are we compliant are we following best practices are we following these standards are we adhering to these policies, you know, they don't have time to worry about the day to day lives of developers making changes to databases.
So, you know, they don't want to be involved in this. So how do the developers and the DBA specifically handle changes to the database when it comes to affecting tables and columns that ultimately are going to start collecting classified information. So, I guess there's two kind of sites for this as well. Like we've been talking a lot about classifying columns, but then there's also the compliance side about, I guess, development, compliant development and bringing compliance into development in other ways and most notably that's going to be source control.
You know, versioning a database like you would your app code means that you track changes and see what gets changed and what and eventually we can start thinking about doing things like locking down development, sorry, not development, testing and production environments to make sure that people don't have access. You know developers should be the ones who are affecting change in production because that means that they've got a level of access they probably should have. So, I think people are great post the other day about someone who basically didn't have access to production to do a lot of changes changes that they needed to do, but did have access to create store procedures. So they basically put everything through store procedures like they would write store procedures to create tables they write store procedures to access data.
It's crazy, absolutely crazy. You know, there's so many different things that you can manage here so yeah, just off to just remove the question entirely. I think something that stands out and I think it's something that you and I talk about an awful lot is that it's process right, you know it's people it's process is tooling, it is part of that whole kind of devops cycle. If a developer makes a change to a table in the development environment, then what we should have is some kind of process that allows them to easily flag that that new column that new table is going to store sensitive information like phone numbers like the needs of birth like names like vehicle registrations.
And that should be part of the development work as much as committing to version controllers as much as handling merge conflicts is all of the parts of these processes should just natively flow together. And if you have a side administrative process like, Oh, by the way, make sure you go and update that Excel spreadsheet. Nobody's going to want to do that nobody is going to do that and it's going to fall apart at some point. But if your deployment of that work hinges on you actually completing the necessary processes, including classifying any new columns.
Then you hit a point that we call ongoing organizational compliance right it is part of the process. So we add a column we tag it with its metadata. That gets built and deployed just the same way as our classifications are just the same way as our actual change would you know hypothetically let's take a red gate component sequel change automation. Yeah, we make a table change generate the migration script source control that migration script we build the database from scratch we deploy it to integration.
So this is where we can now carry out our tests. So we flag the columns being sensitive do a few tests and then the classification alongside the actual deployment of this new column. They just go up from integration to testing to staging to propped Bish bash boss done. And it's all part of the process that means that we persist changes throughout the pipeline.
And it's actually. Simple and straightforward. Exactly. The hard part is actually making sure that you can do the classification in the first place.
Many people think that classifying a database is going to be an easy job. And that's not so not definitely not. And there are two sides to the coin. How do we get classified and then how do we stay classified.
And that's true of anything that's true of you know structured data and unstructured data semi structured data. And it's this is this is playoff of the two where a lot of people get halfway through the first getting getting classified and they go. Oh, this is too much effort I can't be bothered and they give up. Or they don't have the right capabilities at hand you're not going to go through on the in SQL server world you're not going to go through SSMS and you'll go through every single column and update every single every single column.
Again, road by agonizing road with their extended properties saying whether or not they're sensitive, you need something smarter than that you need to be smarter with your approach, both to getting classified and staying classified. And then that way you can feed this effort, you know this effort bridges both compliance and development devs don't feel like compliance impacts their life, because they just do it as part of the process it's really simple. But governance and compliance people and audit people they go, yeah, we don't have to worry because we know that the information we're getting from our development team is up to date. Now what we've got to do is now go and focus on these folk over here who excel sheets into databases and stuff like that but you know that's a separate thing.
So it's a way for these two silos to break down and actually communicate and work more efficiently together. That's the dream right there's a dream that supposedly we're going for. That's fine. All right.
Oh, is it time to move on. I say so. Yes. Basically for now Chris.
It's time for. Let me take a sip of beer. Now the nicest. It's time for the new.
Beautiful Chris beautiful. So let's start with sports. Yeah, go sports. Oh yeah, as everyone.
I'm sure knows it's a huge month for sports at the moment. And the big story on everyone's lips has of course been. The Microsoft Excel financial world cup. Yeah, so yeah, we have a big eight eight eight battle that happened on the 8th of June.
Eight countries took part with eight financial modeling leaders on the 8th to. I can't keep this up. But yeah, I don't know. So kind of financial modeling competition.
As I'm sure you've no doubt heard already. Michael Yarmon from I think it's the army or maybe German. I don't know from Canada one getting 700 points. Sorry if that was a spoiler for anyone.
Yeah, sorry if you had taped the global data modeling championships. I don't even know where to begin with this. But hey, you know, I mean, it's not, I mean, they're using Excel. It's not the Excel championships, but I can imagine they're doing some pretty wizard stuff with Excel for data modeling.
But the very fact that there is a data, a data modeling world championships. I think says to me that we've made it. It warms my heart next month here about Excel artwork. No, seriously though.
Yeah, when it comes to when it comes to sports, it's the big news at the moment is, of course, for all. Yes, with the top story, and everyone's minds being the theft of the FIFA 21 source code from EA. Oh, I was thinking it was more going to be the fact that England were going to use the lose the euros. The what?
The euros. Mate, we never entered that. It's always been pounced. Oh, never mind.
Carry on. I'm just going to cringe over here in a corner. Anyway, anyway, anyway, so EA had about 780 gigs worth of source code, stolen and published online. Well, I say published online being sold online.
Yeah, all very interesting. Did they sell you the first part of the source code and then tell you that you had to pay more money to unlock the rest of it, in which case it was probably an inside job. Yeah, and no doubt we'll see exactly how much of that source code is exactly the same for FIFA 22 as well. The jokes just right themselves, really.
And my final piece of news. So that was as close to a date breach as we're going to get. I'm afraid. Now some of you might have recognized a couple of weeks ago.
Now you had a weird, weird hour where suddenly the Internet stops working. So for around about an hour on, I believe it was Tuesday the was at the eighth of June. Are you forgetting your days again today's Thursday when we're recording? Tuesday the eighth of June.
Yeah, we struggled to access a lot of our favorite websites like Reddit and the Guardian and Amazon and New York Times. Because it's time I spend on the New York Times. I just can't help myself. This is because actually a lot of government websites as well.
This is because of a company who I'm sure you've heard of before called Fastly. So fastly provide something that's called a content delivery network. What this does is essentially allow ISPs to fast load a lot of media content. So Netflix, for example, use something fairly similar to allow ISPs to more quickly transfer video content to rather than having you stream it from Las Vegas data centers.
Yeah. So fastly had a problem that problem was caused by one of their users quite selfishly changing a setting that they were allowed to change. And for whatever reason, that trigger bug and that meant that 85% of the fastly network was down. Which really not.
Yeah, that's absolutely bonkers. And what I guess is crazy here is that like a lot of the Internet is, I guess, kind of unknown to the most people. But as we kind of move forward and things get more sophisticated, more and more organizations have positioned themselves as quite crucial parts to a lot of things. And this is kind of one of those examples.
Yeah, kind of worry because one outage means the whole lot goes down. It's not quite this net of interweaving high availability. We might have worked. Yeah, absolutely.
I think it's just a weird hour. I just remember seeing it on Twitter, like the whole Internet basically being down. Slack and the Microsoft Office 365 suite was still active. So I think it was productive hour.
I think I've had it months. Absolutely. Brilliant. Well, actually, I have a piece of news as well, Chris.
This one comes out of the Redgate camp specifically. And that piece of news is that so I'm sure you're aware, but just in case everyone else isn't aware, Redgate offers a capability called SQL data catalog and SQL data catalog is just kind of what we've been talking about over the past hour or so. It is a system of record for semantic metadata when it comes to columns across tables across everything. And it's really nice.
It's got rest API. It's got a PowerShell module. It's effectively something that you can build into your processes like into a DevOps shameless plug, go to plant based SQL.com and look up on my classification blogs. Because I did it.
I put data catalog into an Azure DevOps deployment process. I did. I did. I did.
I did. I did. I did. I did.
I did. I did. I did. I did.
I did. I did. I did. I did.
I did. I did. I did. I did.
I did. I did. But one thing that's been missing from data catalog for a little while now is that people expect it to scan data. They expect it to just natively be able to look at columns and say, Hey, you know, it looks like you've got a date of birth in here.
It's a very difficult problem to solve because who do you surface this information to? How do you get people to confirm that that's the case? Like what algorithms do you use to use machine learning to use basic pattern matching use combination of things? Do you enable it by default?
There's so many questions, so many things to answer. Well, as of today as of recording the 17th of June 2021, I can confirm that reg eight SQL data catalog now has machine learning and data scanning, or at least the V one of said capabilities, which means that when you point SQL data catalog at your database, it will go off. It will have a quick look at some of the data stored within those tables, and it will try to match it up with a certain, a certain, smaller subset of different types of data, and it will go ahead and say, I think. This is an I ban.
I think this is a first name. I think this that the other. And, you know, this is paving the way. This is huge functionality to get at this stage, and it's paving the way for more kind of configurable.
I'm going to train this model with, you know, first names from France, from Germany, from Slovakia, from Romania, from wherever. But the point is, this V one is an absolutely massive step forward that's going to allow people to really so much easily get classified in the first instance. It's like we said, it's that issue get classified, stay classified, and this is one of those things that's going to help people get classified all that much faster. And hopefully so much easily.
Exactly. And you can go over to the, exactly. And you can go over to the, you can go over to the red gate website that's red hypingate.com. And if you go to products and SQL data catalog, you'll be able to find yourself a download and you can get started with that.
Get yourself a free trial right now. And obviously getting in touch if you're interested in getting your estate classified as well. Request a demo and get Chris something to do it. Oh, get me to show you around data catalog and data masker because I love data masker.
But get me to show you around the magic, the majesty that is data catalog because honestly you haven't lived until you've seen one of my demonstrations. Perfect. Indeed. Well, we've come to the top of the hour, Chris.
Isn't it sad? Top of the hour. I feel like I've talked way too much in this session, but much like the giraffe that doesn't want to let go of its, you collect a scrunch. No wait, that's koalas.
Wait, what is your obviously much like the giraffe that doesn't want to let go of its branch. I think awful much like the giraffe doesn't want to let go of its leaves. I always get a bit of a being my bonnet when it comes to data classification because people don't always do a proper job. For those listening, I just held up my beer bottle, but you've got to do a proper job, make sure you get everything classified and show you've got that ongoing classification as well.
And naturally, the benefits will just be there in their droves for for all time. Sounds great. All right. Well, I think we should end it there.
I'm looking forward to next time where we can make a proper hickery clan caviar and talk too much about the next topic. Whatever that might be. Yes, indeed. Well, stay tuned.
Thank you once again for joining us again. Always reachable. If you want to reach out to send an email to red hyping.com forward slash wait, no DBL at red hyping.com or visit red dash gate.com forward slash DBL or of course, send us a tweet that is at Kurzwell Chris and at plant based sequel on Twitter and we look forward to hearing from you. Please do we get very lonely.
We do get very lonely. We want to hear your ideas. What beer do you want us to drink? What do you want us to talk about?
Pick any subject? Honestly, I'm open to anything and I'll just make Chris follow me along. Well, we've got the next month. Look forward to of research and painstaking endless rehearsal for the next episode of DBL.
Yes, that's a nice one. Something else to do. Absolutely. But for now, it's a thank you very, very much from me.
Cheers. All right. Thank you for Chris. But for now, it's cheers.
Bye.