What's New in the AWS Data and Analytics Stack for 2026 [Webinar]

IT Training and Development Resources
Webinars
What's New in the AWS Data and Analytics Stack for 2026 [Webinar]

Myles will distill and recap the key takeaways from recent announcements, including those from AWS re:Invent 2025 and provide insights into how these changes can impact your data strategy.

The AWS data and analytics stack is constantly evolving, and it can be tough to keep up with the latest changes. In this 1-hour session, Myles Brown, Cloud Solution Architect at Ascendient Learning and highly rated AWS instructor known for his engaging teaching style, guides you through the latest advancements in the AWS data and analytics stack.

Browse our course catalog for AWS Training and Certification or Generative AI on AWS courses. To speak with an AWS expert to determine your best path to certification, contact us.

What's New in the AWS Data and Analytics Stack for 2026

Join Us to Learn About

Key updates to Amazon S3, RDS, Aurora, DynamoDB, Redshift, and other storage and database services.
New features in Glue, Kinesis, Amazon Managed Service for Apache Flink, Quick Sight, Lake Formation, and other analytics services.
The implications of these changes for your business.

At the conclusion, Myles will answer your questions.

Audience

Data engineers
Data scientists
Data analysts
Cloud architects
Anyone who uses AWS data and analytics services

About the presenter:

Myles Brown is a Cloud Solutions Architect at Ascendient Learning. He has over twenty-five years of experience in the IT industry across a variety of platforms. Recognized as an AWS Authorized Instructor Champion and a Google Cloud Authorized Trainer, Myles has delivered award-winning authorized IT training for the biggest cloud providers.

WEBVTT

1
00:00:05.550 --> 00:00:24.620
Anne Fernandez: Hi, everybody! Welcome to today's webinar. It's Webinar Wednesday with Miles Brown. My name is Anne. I'll be here in the background, I'll just do a little intro. I'm here looking at any of the comments, so if you've got any questions that you want to put in the comments field, we'll be taking a look at them.

2
00:00:24.620 --> 00:00:35.239
Anne Fernandez: This session is being recorded, and we will give you the recording after this is all over. Probably sometime tomorrow. We'll go ahead and email you with this and some slides.

3
00:00:35.240 --> 00:00:51.790
Anne Fernandez: But welcome to What's New in the AWS Data and Analytics Stack for 2026. Very exciting, because there's tons of stuff to keep up with, so Miles is kindly going to distill all the stuff that matters down for you in this one hour. He attended AWS reInvent, so you don't have to.

4
00:00:53.260 --> 00:00:54.790
Anne Fernandez: Alright, look…

5
00:00:55.370 --> 00:01:11.209
Anne Fernandez: Take it. So again, just welcome to today's session. Again, my name is Ann, I'm here with Miles, and on behalf of Ascendant Learning, part of Accenture, we are so glad that you're here. Thank you for spending this hour with us. We know you're busy, and we do not take it, like, lightly.

6
00:01:11.210 --> 00:01:20.370
Anne Fernandez: Just in case you don't know who we are, we are an instructor-led IT training company. We have courses that are

7
00:01:20.370 --> 00:01:29.630
Anne Fernandez: live for individuals online. We also have team training and upskilling for your entire enterprise. We do…

8
00:01:29.670 --> 00:01:52.280
Anne Fernandez: certification classes on just the leading certifications and badges, including AWS, Google Cloud, and a lot more. That's there on the right hand of the screen. And we do teach a lot of topics, including AI, ML, data science, cloud, agile DevOps, application development, programming languages, C Sharp, JavaScript, TypeScript.

9
00:01:52.280 --> 00:01:55.910
Anne Fernandez: Business intelligence courses, and… and a lot more.

10
00:01:55.910 --> 00:02:05.150
Anne Fernandez: And just in case you didn't know, we are now Ascendient Learning part of Accenture, which is actually great news for you, because

11
00:02:05.520 --> 00:02:14.920
Anne Fernandez: We still can give you the white glove service that you're used to, but we have Accenture backing us, so let's say you need, consulting with asynchron…

12
00:02:14.920 --> 00:02:30.799
Anne Fernandez: and synchronous training and certifications, we can help you out with everything with the entire life cycle of your training. But we're still us, and we still have our amazing instructors who love what they do. And speaking of amazing instructors who love what they do.

13
00:02:31.510 --> 00:02:48.860
Anne Fernandez: meet Miles Brown, if you haven't already. He's got his email up there, miles.brown at Accenture.com. He is our cloud solutions architect at Ascendient Learning. He's got over 30 years of experience. As you can see, he's got a lot of credentials.

14
00:02:48.860 --> 00:03:00.810
Anne Fernandez: But what I really think makes Miles stand out is his ability to take complex topics and make them very clear and understandable. And actually, Miles is…

15
00:03:00.810 --> 00:03:24.939
Anne Fernandez: doesn't know this, I don't want to embarrass you, Miles, but I was going through a bunch of student evaluations, just, like, looking at quotes and themes, and one theme that came up again and again was Miles' ability to present things in a very clear and systematic way. So it's one thing to know everything, but then to be actually able to explain it, is another. And if you get him on the phone, he's also really good at letting you know,

16
00:03:25.460 --> 00:03:44.059
Anne Fernandez: exactly what your team needs if you… if you… if you tell him what your project is, what your, you know, what your ROI is, your needs, he's really good at pinpointing that. And there's one, I wish I could put it up on the screen, but there was one, eval that I thought was amazing and cracked me up. It said, great job, Miles.

17
00:03:44.060 --> 00:03:50.180
Anne Fernandez: inspiring, even, so… There you go, you're gonna have a very inspiring session today, I'm sure.

18
00:03:50.180 --> 00:03:50.600
Myles Brown: I've.

19
00:03:50.960 --> 00:03:53.300
Myles Brown: But you've really built it up now.

20
00:03:53.300 --> 00:04:18.289
Anne Fernandez: Yeah, that's right. You have to be inspiring, too. All right, so with that, I will be here to answer any questions, you know, just, like, broadly about the business and about training, but if you've got questions, like, about AWS, I think Miles will probably be the one to have to answer them, and you can go ahead and put them in the chat. Okay, so I will… I'll turn my camera off, and I'll mute myself, but

21
00:04:18.290 --> 00:04:19.579
Anne Fernandez: I will be here.

22
00:04:19.810 --> 00:04:21.440
Myles Brown: Alright, thanks, Miles!

23
00:04:21.579 --> 00:04:22.369
Myles Brown: Thanks, Anne.

24
00:04:23.689 --> 00:04:30.919
Myles Brown: Yeah, so I have been an AWS instructor since 2014. What's that? It's coming up in 12 years, I guess? Wow.

25
00:04:31.159 --> 00:04:43.249
Myles Brown: And I also teach some Google Cloud and some Oracle Cloud. I included a couple of, certifications here that are, they're still…

26
00:04:43.579 --> 00:05:00.159
Myles Brown: I still have them, they're still, they haven't expired yet, but I can't get them again, because those two certifications are going away, or have gone away. So there's no more database specialty, there's no more data analytics specialty. And so at the very end, I'll talk about some of the certification changes.

27
00:05:00.389 --> 00:05:11.389
Myles Brown: That AWS has made. I also have this one from Oracle, which is the Oracle database at AWS, architect one. That… it is, you know.

28
00:05:11.879 --> 00:05:26.819
Myles Brown: it is sort of a partnership that AWS and Oracle have between their two clouds. It's something that became generally available this year, into 2025. They announced it, you know, well before.

29
00:05:26.819 --> 00:05:34.779
Myles Brown: And Oracle sort of has done this for a little while. I'm going to talk about it now because I didn't include it in the, in the presentation, but I thought, oh, I should talk about that.

30
00:05:34.899 --> 00:05:44.619
Myles Brown: You know, Oracle was making it more and more difficult, more expensive to run Oracle Database in AWS, in Azure, in Google Cloud.

31
00:05:45.029 --> 00:06:04.509
Myles Brown: And that really drove some of those cloud vendors to try and look for alternatives, right? And so with AWS, you know, they created Aurora to say, you know, we can get the same kind of scale and throughput as the commercial, you know, RDS vendors. So, like, Oracle and SQL Server.

32
00:06:04.519 --> 00:06:20.139
Myles Brown: But at, you know, maybe a tenth of the cost. And so that took a lot of the business away from Oracle, you know, people said, okay, I can see a path to getting off of Oracle Database. But there's other people who say, you know, I'm too ingrained with Oracle, I can't do that.

33
00:06:20.539 --> 00:06:29.009
Myles Brown: And so, you know, I want to use the cloud, but you're not making it easy for us. And so, Oracle kind of said, well, if you can't beat them, join them, right?

34
00:06:29.029 --> 00:06:44.929
Myles Brown: So Oracle has its own cloud, OCI. It's not terribly popular, like, it's well down the list of, you know, like, it's AWS, Azure, R1, 2, and then a, you know, distant third place is Google, and then way down the list is Oracle.

35
00:06:44.969 --> 00:06:59.079
Myles Brown: But Oracle says, well, what you could do is, you know, we've got places co-located, so if you want to run your infrastructure in AWS, but run your Oracle database in OCI, we can connect them together nice and easily.

36
00:06:59.079 --> 00:07:06.049
Myles Brown: So that's what they call Oracle Database at AWS. And they have a similar thing for Azure and for Google Cloud as well.

37
00:07:06.599 --> 00:07:13.239
Myles Brown: And so that's, that's kind of a new thing that, just this year, the AWS one finally took off.

38
00:07:13.789 --> 00:07:24.029
Myles Brown: So we're not going to talk about that in the official course, I just wanted to mention that. Here's… here's my agenda for today. I'm going to start with kind of the big picture, and then look at one of the biggest

39
00:07:24.229 --> 00:07:39.039
Myles Brown: sort of new announcements, I guess? Not a big one for most people, but we'll talk about QuickSuite. And then, I think kind of the two big things we're going to talk about are Iceberg and Zero ETL, and then there's a bunch of small things to talk about.

40
00:07:39.799 --> 00:07:45.529
Myles Brown: And then just, you know, AWS certification changes. And we'll have some time for Q&A at the end.

41
00:07:45.659 --> 00:07:53.789
Myles Brown: Along the way, if you have questions, throw them in the chat. I'm looking at the chat, you know, maybe just pull the pull down to say everyone, so everybody can see the question.

42
00:07:54.499 --> 00:08:09.879
Myles Brown: And that'll be the easiest way. So, big picture, what we saw in the data and analytics stack, we're not seeing a lot of new services over the last year. So, this is a presentation I do every January, you know, basically looking at what's new from

43
00:08:09.879 --> 00:08:19.619
Myles Brown: the last year, especially the last, like, month, right? Because every year, AWS has their big reInvent conference, the first week of December, and

44
00:08:19.619 --> 00:08:27.919
Myles Brown: for the week leading up to it, and maybe that week of that, and the week after that, there's a lot of new announcements, and so we're kind of… I try and…

45
00:08:27.999 --> 00:08:37.509
Myles Brown: you know, now there's literally thousands of announcements, so what I do is I say, well, just the ones in the data and analytics stack, let's just look at those.

46
00:08:37.599 --> 00:08:45.359
Myles Brown: And this year, even over the whole year, there's not that many new, services. It's a lot of little features and stuff.

47
00:08:45.389 --> 00:08:53.659
Myles Brown: Where you'll see the most movement is in the AI and ML space, especially generative AI, agentic AI stuff. And so.

48
00:08:53.659 --> 00:09:08.359
Myles Brown: I'm gonna leave that out of the scope of this presentation, and I think we'll do a separate presentation just on, you know, what's the state of generative AI at AWS. I'm thinking that might be a month or two from now, we'll do another one of those, just to kind of…

49
00:09:08.529 --> 00:09:26.859
Myles Brown: we have some new courses that just came out that, like, they're coming out next week on Agentic AI, on AWS courses, so, you know, there's a lot of activity in that area. So what they did with the data and analytics stack, the major focus is really fulfilling the vision of a bunch of announcements they made a year ago.

50
00:09:26.889 --> 00:09:39.899
Myles Brown: Right? So they announce it, but it takes them 6 months before it becomes GA, and then they've got a bunch of another announcements to fill it out and make it work. And so, that's really what this year was.

51
00:09:39.899 --> 00:09:47.229
Myles Brown: Now, the one big announcement, I would say, was, this, the, introducing Amazon Quick Suite.

52
00:09:47.429 --> 00:09:51.509
Myles Brown: Now, you may have heard of QuickSite, which is the business intelligence tool.

53
00:09:52.019 --> 00:10:02.289
Myles Brown: And, you're probably not using it, because most people aren't. Like, it's not the most popular. You know, when you look at business intelligence tools, the big ones are Tableau.

54
00:10:02.309 --> 00:10:16.119
Myles Brown: Microsoft Power BI, and then there's a bunch of, you know, Cognos and, like, other ones. But, and so, which one you're using within a large organization probably is mostly determined by

55
00:10:16.119 --> 00:10:35.379
Myles Brown: you know, were you an IBM shop at one point? Were you in a Microsoft shop? Like, that… you know, if you're mostly using Microsoft stuff, you're probably going to use Power BI, right? But, Tableau was, you know, otherwise the biggest one, but there's a whole bunch of other ones, like Qlik and Domo, and, you know, the list, there's probably 40 others, right? And so, really.

56
00:10:35.379 --> 00:10:48.819
Myles Brown: depending when you made a big investment into business intelligence, that probably was, you know, whatever tool just came out with the best bells and whistles at that time. And so, AWS introduced

57
00:10:48.819 --> 00:10:56.169
Myles Brown: something called Quicksite. And they did this probably, I want to say, like, 6 years ago, 7 years ago?

58
00:10:56.179 --> 00:11:03.439
Myles Brown: And, it's sort of… Interesting, because AWS, when they build those, like, end-user tools.

59
00:11:03.629 --> 00:11:07.889
Myles Brown: They kind of… they put that minimal viable product out there.

60
00:11:07.889 --> 00:11:26.579
Myles Brown: And they really put the minimal in minimal viable product, right? And so a lot of people, they're like, oh, AWS has a business intelligence tool, let's go check it out. And you look at it when it's in preview, or even once it's GA, and what you'll find is that, oh, it looks a lot like Tableau, but it doesn't have all the bells and whistles.

61
00:11:27.029 --> 00:11:29.809
Myles Brown: And so, it's not very impressive when you first look at it.

62
00:11:29.819 --> 00:11:47.509
Myles Brown: But what they do is they put it out there, they get it in use in some big customers, and then they get the feedback, and then they say, what else do you need? What else do you need? They keep adding to it, right? And so, you know, if you never looked at Quicksite for the last, you know, few years, come back and look at it. There's some really cool stuff in it.

63
00:11:47.649 --> 00:11:59.789
Myles Brown: But, once, generative AI started making its way into all the other tools, and same thing with Quicksite, it had Amazon Q built into it.

64
00:11:59.959 --> 00:12:06.609
Myles Brown: they decided to completely rebrand it and kind of redo it from the ground up. And so that's what they introduced

65
00:12:06.769 --> 00:12:08.739
Myles Brown: Just about 2 months ago.

66
00:12:08.969 --> 00:12:13.849
Myles Brown: So QuickSuite is the Agentic teammate that quickly answers your questions

67
00:12:13.949 --> 00:12:30.509
Myles Brown: your questions at work, and turns those insights into actions for you. So, it brings, like, AI-powered research, BI, automation capabilities all into a single workspace. So it starts with the basic concepts of QuickSight, like the BI tool.

68
00:12:30.579 --> 00:12:40.759
Myles Brown: But it's got this idea of quick research, quick flows, quick automate, and so that's… that's the big idea. Let me see, I've got it here somewhere…

69
00:12:42.249 --> 00:12:48.439
Myles Brown: Okay, so if I just go to my console and I start looking up… quick…

70
00:12:51.479 --> 00:12:52.359
Myles Brown: Sweet.

71
00:12:53.129 --> 00:12:54.149
Myles Brown: There it is.

72
00:12:54.869 --> 00:13:05.599
Myles Brown: Now, it's one of those things where you have to go and set up Identity Center and set up users inside of there, because, you know, it doesn't just use

73
00:13:05.859 --> 00:13:08.809
Myles Brown: The users in your,

74
00:13:08.949 --> 00:13:19.569
Myles Brown: you know, that you would have set up for using console or something like that, because it's probably different people. It's a bunch of business analysts that are going to use the BI tool and the whole quick suite.

75
00:13:19.609 --> 00:13:32.809
Myles Brown: As opposed to the people who you probably already have set up with AWS accounts inside your AWS account, right? So, you sort of set it up, and you, you know, just give them an email and a password, and that lets them get in there.

76
00:13:33.189 --> 00:13:46.499
Myles Brown: And so you come in, you start with data sets, so here I uploaded just 10,000 customer CSV file, and then you start doing some analysis. So here's one I started already, I told it to go and grab.

77
00:13:46.889 --> 00:13:51.919
Myles Brown: customers by city and country, and so I've got this kind of…

78
00:13:52.239 --> 00:13:58.579
Myles Brown: you know, came up with this pie graph or whatever. Oops, where am I going there? Come back.

79
00:14:00.119 --> 00:14:01.079
Myles Brown: Quick sweep.

80
00:14:03.349 --> 00:14:04.529
Myles Brown: What's gonna happen here?

81
00:14:07.069 --> 00:14:08.259
Myles Brown: I lost it.

82
00:14:13.749 --> 00:14:14.649
Myles Brown: Quebec.

83
00:14:15.229 --> 00:14:18.329
Myles Brown: Alright, now we're back.

84
00:14:18.459 --> 00:14:37.799
Myles Brown: So I, you know, I can build dashboards and things like that. And then here, you'll see you can do automations, enhance your dashboards. So they have this concept called scenarios, so I can come in and create a scenario. You can upload data, you know, just from scratch, or if I already have a dashboard with some data.

85
00:14:37.799 --> 00:14:40.069
Myles Brown: Let me just grab this dashboard.

86
00:14:40.889 --> 00:14:42.929
Myles Brown: And just add all the data.

87
00:14:43.889 --> 00:14:52.489
Myles Brown: and say, okay, I want to start some analysis. It says, okay, we'll start with a problem you're trying to solve. Which city

88
00:14:53.269 --> 00:14:59.429
Myles Brown: I don't know, let's do this. Which country has the most… Customers.

89
00:15:00.869 --> 00:15:01.589
Myles Brown: Sure.

90
00:15:01.819 --> 00:15:02.999
Myles Brown: Something like that.

91
00:15:03.159 --> 00:15:18.369
Myles Brown: And start the analysis. So it's gonna go and… it starts grabbing that data. Now, this one, I'm just grabbing the data that was used underlying that one dashboard, but I can upload more data right in here while at least sort of like a scratch pad.

92
00:15:18.499 --> 00:15:23.129
Myles Brown: Where I can, you know, move from data to insights. And I don't have to go in

93
00:15:23.219 --> 00:15:35.679
Myles Brown: you know, set up all the data sources, pull them all together. If I just have, like, an Excel spreadsheet, and I want to pull that in and kind of add that to my scenarios, and say, well, what if, you know, we add these three cities, you know, what would happen?

94
00:15:35.759 --> 00:15:53.409
Myles Brown: And so it says, okay, based on the data that you pulled in, and this concept of, you know, most customers or whatever, it says, here's a bunch of questions we could ask, show the top 5 countries with the most customers, list the cities with the highest number of customers in Brazil, you know, just giving you examples

95
00:15:54.029 --> 00:15:57.379
Myles Brown: Top 5 countries with the most customers.

96
00:15:58.259 --> 00:16:09.319
Myles Brown: Sure, I'll do that one. And so it's gonna go and grab that data, probably show it to me in a tabular form, but also then go and pick some sort of visualization for it, too.

97
00:16:09.509 --> 00:16:17.949
Myles Brown: And so you're asking questions in, like, plain English, and it's grabbing that data, and it's going and saying, okay, well, Korea, Congo, I know how to…

98
00:16:18.149 --> 00:16:22.919
Myles Brown: Not really? That's kind of a weird collection. Oh, and now it's building…

99
00:16:23.179 --> 00:16:26.759
Myles Brown: I think it's building me a horizontal bar chart.

100
00:16:26.869 --> 00:16:27.979
Myles Brown: By default.

101
00:16:28.469 --> 00:16:30.929
Myles Brown: With the 5 most, right?

102
00:16:31.239 --> 00:16:44.269
Myles Brown: And so that's the idea, is that you can keep going back and saying, okay, I'd like to add some more stuff, what if we expanded our operations to Vietnam? How many potential customers could we reach, you know, with the current data, you know. So it's just…

103
00:16:44.269 --> 00:16:59.589
Myles Brown: spitballing based on that first question you asked. Here's some other questions. So, if you're just trying to, like, you know, play around with the data, it gives you some inspiration, but, you know, as you ask more questions, it'll come up with similar other questions around the data.

104
00:17:00.239 --> 00:17:09.179
Myles Brown: So that's the scenario idea, but like I said, you can go through and, you know, set up flows and all kinds of stuff.

105
00:17:10.029 --> 00:17:20.509
Myles Brown: I think here in the slides, I mentioned, the flows and Quick Automate help you, automate business process from simple routine tasks to complex multi-department workflows.

106
00:17:20.709 --> 00:17:25.639
Myles Brown: And if you go read the announcement, they kind of explain some examples of all that.

107
00:17:25.889 --> 00:17:33.709
Myles Brown: So this was announced in… oh, in October, so about 3 months ago now is when QuickSuite first sort of came out.

108
00:17:33.869 --> 00:17:40.669
Myles Brown: So that was kind of the one big new service, although most of that service is QuickSight, which isn't even a new service.

109
00:17:41.329 --> 00:17:49.549
Myles Brown: I would say the number two, kind of, big change in the analytics stack is how much they're embracing Iceberg.

110
00:17:49.669 --> 00:17:54.299
Myles Brown: If you're not familiar with Apache Iceberg, it's an open table format.

111
00:17:54.719 --> 00:18:00.369
Myles Brown: There's been various support for it at AWS since 2021.

112
00:18:00.779 --> 00:18:08.859
Myles Brown: But really, I would say this past year, what they did is they said, we're making Iceberg the center of our universe as far as storing data.

113
00:18:09.479 --> 00:18:26.519
Myles Brown: And they're effectively removing the wall between data lakes and data warehouses, right? So, data warehouses have been around since the 90s, it's that idea of, you know, it's a traditional relational database, albeit one that's, you know, tuned a little differently than your transactional databases.

114
00:18:26.519 --> 00:18:34.139
Myles Brown: But, you know, the big idea would be you would normally take all your data from your transactional places, and you would

115
00:18:34.269 --> 00:18:46.269
Myles Brown: at least back in the 90s, we would nightly, you know, do an ETL job, where we would extract data from all kinds of places, transform it a little bit, you know, aggregate it, probably, and then load it into a data warehouse.

116
00:18:46.899 --> 00:18:56.709
Myles Brown: And that data warehouse was really designed not for single-row insert updates and deletes, but for bulk inserts and then huge queries all day long, right?

117
00:18:56.989 --> 00:18:59.759
Myles Brown: And then, you know, because…

118
00:19:00.029 --> 00:19:06.739
Myles Brown: It's kind of an expensive place to store data, and it's really good at querying really big sets of data.

119
00:19:07.109 --> 00:19:07.949
Myles Brown: But…

120
00:19:08.559 --> 00:19:21.369
Myles Brown: it is a relational database, meaning your data has to kind of fit into the rows and columns of a relational database table, right? And so, once, you know, Hadoop came along in the late

121
00:19:21.459 --> 00:19:35.229
Myles Brown: you know, 2006 to 2009 kind of time is when, you know, big data started to take off, and people started to build this concept of a data lake. And the cloud really enabled that a lot, because we had a good, cheap.

122
00:19:35.359 --> 00:19:48.929
Myles Brown: storage place to hold all of our data across our enterprise, and that's S3, at least in AWS. And so that idea of a data lake where we have a central place where we store all the data in its natural format.

123
00:19:49.539 --> 00:20:08.119
Myles Brown: And we could use it for all kinds of things. I could take a subset of that, clean it up, and put it into a data warehouse. Or, if I'm a data scientist, I could go in, look at that raw data, right? Because as soon as you start processing it, you're going to lose some context, right? And so I may want to, you know, access that raw data.

124
00:20:08.239 --> 00:20:18.789
Myles Brown: And so, the concept of data lakes have been wildly popular for probably the last 7, 8 years now. People have been doing a lot of that.

125
00:20:19.199 --> 00:20:20.079
Myles Brown: Well…

126
00:20:21.329 --> 00:20:40.749
Myles Brown: The problem with S3 is it's really designed kind of with the worm model in mind, where you write once, you read many times, right? And so, if you think of a data lake as, well, this is just historical data, I grab it, I throw it in there, that data doesn't change. If you want to, you know, change the data, you might make

127
00:20:40.809 --> 00:20:52.639
Myles Brown: Like, different phases within your data lake, where you've got the raw data, and then you take it, and you do some sort of cleanup on it, put it into another sort of, like, phase of my data lake.

128
00:20:52.639 --> 00:21:11.389
Myles Brown: Some people have, like, bronze, silver, gold, or, you know, the different, like, a landing zone, and process data, or, you know, gold data is for, you know, maybe even it's… it's been stripped of personally identifiable information or something, so that other people can query it.

129
00:21:11.389 --> 00:21:14.649
Myles Brown: Right? So, that's kind of the idea, but…

130
00:21:14.649 --> 00:21:29.899
Myles Brown: we do run into some problems with just using S3 for data lakes, because it's not a relational database. It's not set up with those kind of ACID-compliant transactions. And these days, what we find is we've got raw data sitting there, but at some point.

131
00:21:29.899 --> 00:21:35.549
Myles Brown: you know, somebody from Europe comes along and says, I want you to remove my data, right? They've got these…

132
00:21:35.799 --> 00:21:41.839
Myles Brown: data sovereignty laws, these data privacy laws that say, I want you to remove all your data about me.

133
00:21:42.079 --> 00:21:49.859
Myles Brown: I have to be able to go and find it and delete it, right? And so deleting, like, individual rows and things like that is very difficult.

134
00:21:50.079 --> 00:21:58.769
Myles Brown: With just S3. And so, Iceberg was this, open table format that would give us those kind of capabilities.

135
00:21:58.839 --> 00:22:10.279
Myles Brown: And so, if you just want to know more about Iceberg, this is a pretty good page that AWS has. It just talks about what is Iceberg, and what are the benefits that it gives us.

136
00:22:10.279 --> 00:22:19.379
Myles Brown: And it gives us, essentially, a transactional data leak. So those ACID-compliant transactions, atomic, consistent, isolated, durable.

137
00:22:19.679 --> 00:22:32.989
Myles Brown: And it gives us lots of, you know, we can use the regular kind of SQL syntax. We have the data consistency. Data versioning is an interesting one, so I can go and say, run this query as of this date.

138
00:22:33.169 --> 00:22:38.949
Myles Brown: Because we've got versions on the data, we can look at the data at a certain point in time, even.

139
00:22:39.019 --> 00:22:55.319
Myles Brown: So there's some pretty cool stuff that Iceberg does for us. And so Iceberg, like I said, it's been around a little bit in AWS. You could… you could query Iceberg tables using Athena, right? The big announcement a year ago

140
00:22:55.319 --> 00:23:04.949
Myles Brown: was this concept of S3 tables, and so they've really embraced it using S3, a new type of bucket that is set up for iceberg tables.

141
00:23:04.949 --> 00:23:16.579
Myles Brown: But this year, in 2025, here's some of the major announcements. We now have support for Iceberg version 3, and that gave us some kind of cool new features, some deletion vectors, it's just a lot…

142
00:23:16.579 --> 00:23:26.339
Myles Brown: A lot faster at row-level updates. And row lineage, so precise change tracking at the row level with auditability and everything else.

143
00:23:26.689 --> 00:23:34.399
Myles Brown: In Glue, we now have this idea of materialized views that are sort of managed tables in iceberg format.

144
00:23:34.929 --> 00:23:39.829
Myles Brown: And they get automatically kept up to date as the source data changes.

145
00:23:40.769 --> 00:23:56.439
Myles Brown: We can even query remote iceberg catalogs, so if you've got Snowflake, or Confluent, or, you know, there's a few different places where Iceberg's been, you know, it's much bigger than just AWS, right? Like, a lot of people are embracing this idea.

146
00:23:56.439 --> 00:24:04.269
Myles Brown: And so now, I can kind of, in glue query tables from those external catalogs as long… that are iceberg tables as well.

147
00:24:04.419 --> 00:24:10.459
Myles Brown: So I don't have to move… I don't have to, like, literally move the data from one place in AWS to then query it.

148
00:24:11.249 --> 00:24:22.779
Myles Brown: We even have, support now, this is a fairly new one, write support in Redshift. So, Redshift, you can run a query and directly write that data to an iceberg table in S3.

149
00:24:23.079 --> 00:24:34.329
Myles Brown: And so there's, you know, this idea of being able to move data from the data lake into Redshift, and from Redshift into the data lake, and if the data lake's using Iceberg, everything's still cool.

150
00:24:34.449 --> 00:24:40.519
Myles Brown: And so it's just that overall, you know, support for iceberg format across the board.

151
00:24:40.739 --> 00:24:55.179
Myles Brown: And then specifically, you know, with S3 tables that came out, like, a little over a year ago, in 2025, it transitioned from the initial launch to a really mature, globally available feature set.

152
00:24:55.179 --> 00:25:11.019
Myles Brown: So here's some of the announcements in 2025. They now have intelligent tiering, so just like other S3, buckets, where you don't have to know exactly, like, okay, I want it in standard for a while, and then move it to this, and move it to that, I can use intelligent tiering.

153
00:25:11.579 --> 00:25:17.739
Myles Brown: We have, automatic cross-region replication of iceberg tables for disaster recovery kind of thing.

154
00:25:18.089 --> 00:25:28.299
Myles Brown: Direct query support. So using Athena, I can, right there in the, in the S3 console, I can query S3 tables. That's cool.

155
00:25:28.979 --> 00:25:35.669
Myles Brown: Support for some new types of compaction to really improve query performance on those iceberg tables.

156
00:25:36.359 --> 00:25:49.149
Myles Brown: S3 tables now generate CloudTrail events, you know, so it was really just expanding something that they announced a year ago, and was like, okay, here it is, it's an idea, now you can really use it in production.

157
00:25:49.739 --> 00:25:59.319
Myles Brown: And the last one here is kind of exciting. I can use tagging and, you know, what we would call attribute-based access control.

158
00:26:00.899 --> 00:26:11.299
Myles Brown: To, to really, you know, have these tags on different things, and then use those to say who's allowed to read it, who's allowed to delete it, you know, things like that.

159
00:26:11.559 --> 00:26:17.859
Myles Brown: Rather than trying to use, like, the S3 prefixes and the regular IAM kind of…

160
00:26:18.149 --> 00:26:22.059
Myles Brown: permissions, I can… I can use more of a… an ABAC.

161
00:26:22.329 --> 00:26:37.389
Myles Brown: And then a whole bunch of other announcements that were just sort of global expansion. Now it's in more regions, now it's in more regions, you know. And so that tells you a little bit about how important Iceberg is across, you know, this… this idea of the…

162
00:26:37.469 --> 00:26:51.369
Myles Brown: not just S3, but Redshift, Glue, Athena, they all support it in a big way. And so that's kind of, I would say, the major shift this year was to say Iceberg is our

163
00:26:51.609 --> 00:26:54.199
Myles Brown: Open data format of choice.

164
00:26:55.989 --> 00:27:15.089
Myles Brown: Another thing that's been around for a little bit, but then started to expand again this year, is the concept of zero ETL. And zero ETL, I mean, we talked about what ETL is, it's that data, you know, taking that data from all kinds of places, extracting it, transforming it, loading it into, say, a data warehouse.

165
00:27:15.989 --> 00:27:21.449
Myles Brown: Well, zero ETL is a fundamental shift in how we approach that data integration.

166
00:27:21.609 --> 00:27:34.689
Myles Brown: And it's the idea that we should be able to do sort of an automated, fully managed data replication from an AWS service, like an RDS database, into a data warehouse or a data lake.

167
00:27:34.839 --> 00:27:41.739
Myles Brown: Without requiring a whole bunch of pipeline development that I, you know, that data engineering pipelines that we build.

168
00:27:41.949 --> 00:27:58.629
Myles Brown: So it was first announced in 2023, and they steadily add more and more to it. This year, they now have integration in SageMaker. We're not going to talk a lot about SageMaker, because, you know, like I said, we're going to separate and talk about the AIML world.

169
00:27:58.699 --> 00:28:10.179
Myles Brown: In another session. But, you know, I can now easily access my Aurora MySQL and RDS data for MySQL in SageMaker.

170
00:28:10.179 --> 00:28:25.329
Myles Brown: Glue now supports zero ETL integrations from DynamoDB and 8 applications. So, when we talk about 8 applications, what we're talking about is third-party, applications, like Salesforce, SAP, ServiceNow, Zendesk.

171
00:28:25.649 --> 00:28:35.549
Myles Brown: Right? So, if you've got connections to those things, you can have that data, you know, brought directly into,

172
00:28:36.059 --> 00:28:47.959
Myles Brown: into your data warehouse, into your data lake, they make it really easy. Now, their initial SAP, was… it wasn't everything. Did they mention here what it is?

173
00:28:48.629 --> 00:28:56.729
Myles Brown: I think there was only certain kinds of entities, and then just the last month, they had added additional SAP entities.

174
00:28:56.859 --> 00:29:01.999
Myles Brown: And so, if you're an SAP person, you probably recognize some of these things.

175
00:29:02.489 --> 00:29:06.079
Myles Brown: Let's see, where is it?

176
00:29:10.989 --> 00:29:11.799
Myles Brown: Yeah.

177
00:29:13.269 --> 00:29:20.029
Myles Brown: Entities that lack deletion tracking flags? Yeah, okay. Anyway, so that's a very niche thing.

178
00:29:20.659 --> 00:29:33.999
Myles Brown: The biggest announcement, I think, at reInvent this year around zero ETL was this one, announcing Glue Zero ETL for self-managed database sources. So that's like an on-prem database that you have.

179
00:29:34.689 --> 00:29:42.509
Myles Brown: As long as it's, like, Oracle, SQL Server, MySQL, or Postgres, you can replicate data from that.

180
00:29:42.729 --> 00:29:47.019
Myles Brown: from an on-prem or from EC2 into Redshift, very simple.

181
00:29:47.379 --> 00:29:52.789
Myles Brown: Right? And they make your life really easy for setting up that… that data and moving in.

182
00:29:52.959 --> 00:29:53.689
Myles Brown: Right?

183
00:29:53.829 --> 00:30:05.319
Myles Brown: So, zero ETL, expanding, you know, saying, maybe we can make the role of the data engineer who's traditionally built these pipelines, make it a little easier, hook things up.

184
00:30:06.799 --> 00:30:13.419
Myles Brown: And not have to go and schedule this job nightly to do this movement of data, that kind of thing.

185
00:30:14.099 --> 00:30:16.709
Myles Brown: We have a few,

186
00:30:17.069 --> 00:30:30.729
Myles Brown: A few things that fall under, sort of, the data integration and governance umbrella. So, some new tools were launched to help data engineers. If you are building those complex workflows and dealing with things like multi-account.

187
00:30:30.759 --> 00:30:47.409
Myles Brown: Right? They… there's some… some things that help you here. First off, they now support GLU 5.0, and it supports Spark 3.5. If you're familiar with that. The big thing it does there is native fine-grained access control through lake formation.

188
00:30:48.049 --> 00:30:55.009
Myles Brown: So it really allows us to do row and cell level security on an S3 data lake. So you can read about how that works.

189
00:30:55.209 --> 00:31:01.799
Myles Brown: And so that glue catalog keeps track of, you know, what's going on, and we can have permissions at…

190
00:31:01.919 --> 00:31:04.829
Myles Brown: At the, row level, at the cell level.

191
00:31:05.849 --> 00:31:19.629
Myles Brown: The idea of AWS Clean Rooms now has cross-region collaboration, so if you're not familiar with Clean Rooms, it's this idea where, like, say you have multiple companies that each have some proprietary data.

192
00:31:19.859 --> 00:31:21.779
Myles Brown: That they don't want…

193
00:31:21.989 --> 00:31:29.799
Myles Brown: they want to share parts of it with other people, but not all of it, right? What they… what they want to do is say, okay, we can make this

194
00:31:29.959 --> 00:31:43.049
Myles Brown: place where we can each dump some data. Nobody's allowed to query the individual data, have access to it, but you can query the shared data, right? That's the idea of a clean room. And so now.

195
00:31:43.219 --> 00:31:53.289
Myles Brown: they can do that across regions, or even across Snowflake regions, without having to move or copy the data, so it's the multi-region access is the thing they added.

196
00:31:54.149 --> 00:32:05.209
Myles Brown: Speaking of, if you've got, you know, multiple data warehouses, you know, there is a lot of customers that are adopting multi-warehouse architectures.

197
00:32:07.149 --> 00:32:20.279
Myles Brown: With Redshift, they now have this concept of federated permissions. So what you do is you define the permissions once from any of your Redshift warehouses, and then it automatically enforces that across all the warehouses in the account.

198
00:32:20.639 --> 00:32:23.819
Myles Brown: And so it just sort of makes your life a little easier.

199
00:32:23.819 --> 00:32:41.269
Myles Brown: There's a bunch of other Redshift announcements. Again, this idea of data shares, where I have my data and I want to share it, you know, maybe with another account or something like that. We now have the idea of materialized views on there, and they added some features to that.

200
00:32:41.879 --> 00:32:53.429
Myles Brown: The general availability of multidimensional data layouts, MDDL, we sometimes call that. Here's the big idea behind that.

201
00:32:54.789 --> 00:33:08.769
Myles Brown: Yeah, sort data based on actual query filters, accelerating query performance. So, unlike traditional sorting methods that sort data based on fixed columns, you can do it based on query filters. So, sales in the US

202
00:33:08.969 --> 00:33:14.749
Myles Brown: So, up to 10 times better performance compared to using only single column sort keys.

203
00:33:15.159 --> 00:33:20.059
Myles Brown: And so that's… that's sort of a new thing that they, they brought out.

204
00:33:20.219 --> 00:33:24.229
Myles Brown: Oh, we now have support for the super data type.

205
00:33:24.619 --> 00:33:33.589
Myles Brown: For databases that are case insensitive collation, so this is where you want to use, like, JSON, basically.

206
00:33:33.889 --> 00:33:37.509
Myles Brown: As… as a data type kind of thing.

207
00:33:38.099 --> 00:33:44.479
Myles Brown: Explore for that in Redshift. It now enables cluster relocation by default.

208
00:33:44.929 --> 00:33:49.999
Myles Brown: So the big idea there is if,

209
00:33:50.839 --> 00:34:01.569
Myles Brown: if you have, say, an availability zone goes down. You know, AWS never likes to talk about it, but, you know, when it happens, we all know about it.

210
00:34:01.639 --> 00:34:12.579
Myles Brown: Basically, you can move your cluster to another availability zone when the resource constraints disrupt cluster operations, and that's turned on now by default.

211
00:34:12.699 --> 00:34:16.249
Myles Brown: Okay, so when you create an Amazon Redshift cluster.

212
00:34:16.369 --> 00:34:24.699
Myles Brown: or restore a cluster from Snapshot, it now has that turned on by default. So… You know.

213
00:34:24.819 --> 00:34:37.249
Myles Brown: It just adds to the resiliency. You know, we used to have just this, the automatic detection and recovery from drive and node failures. So this adds another level against AZ-level instance problems, right?

214
00:34:38.149 --> 00:34:41.359
Myles Brown: So that's, that's a few new things on Redshift.

215
00:34:42.319 --> 00:34:51.169
Myles Brown: If we look at, storage, we'll get to S3, let's start with Kinesis. Kinesis data streams, they, they…

216
00:34:51.509 --> 00:35:00.359
Myles Brown: The maximum record size was 1MB, now it's 10, you know, so larger, more complex streaming events can be processed without chunking.

217
00:35:00.579 --> 00:35:19.489
Myles Brown: If you wanted to do that before, you would have to use, you know, something else like Kafka or something. Managed streams for Kafka can have much bigger payloads, but… but Kinesis, which is much easier to use and integrates nicely with all the other AWS services, you know, really had a problem with the max payload size.

218
00:35:21.739 --> 00:35:31.589
Myles Brown: Speaking of size increases, the biggest thing, I think, here is S3, increased the object size. Was 5TB, now it's 50 terabytes.

219
00:35:31.919 --> 00:35:46.119
Myles Brown: So, simplifies the processing of large objects like high-resolution videos, seismic data files, AI training data sets, and more, right? So that 5TB limit had been there since, I believe.

220
00:35:46.279 --> 00:36:00.509
Myles Brown: since 2006, when S3 started, right? And I would teach classes, and I would tell people, like, eventually, we're gonna have to make that bigger when we get to, like, I don't know, 8K video or whatever. But, you know, we're there, and so they've done it.

221
00:36:00.649 --> 00:36:09.229
Myles Brown: They made some improvements to storage lens, you know, so you can get more information, deeper insight into your storage performance and usage patterns.

222
00:36:09.689 --> 00:36:27.459
Myles Brown: Some other stuff with S3, we now have conditional deletes in S3, general-purpose buckets, attribute-based access control, we talked about the ABAC with S3 tables. Well, first they introduced it with regular S3 buckets, then now with the S3 tables as well.

223
00:36:27.939 --> 00:36:34.779
Myles Brown: S3 introduced a new way to verify the content of stored datasets. This is sort of,

224
00:36:35.379 --> 00:36:39.649
Myles Brown: you know, SWI internally does some verification

225
00:36:39.739 --> 00:36:58.569
Myles Brown: of data. You know, when you move data, it does some checksums to make sure, you know, the data that was there is now the same as the data over here, or whatever. Well, now, you can use S3 batch operations, and you can compute your own checksum, job.

226
00:36:58.689 --> 00:37:13.149
Myles Brown: And so you provide a list of objects, a manifest, or specify the bucket with filters, and then you choose Compute Checksum as the operation type, and tell it which algorithm to use, and when the job completes, you get a detailed report.

227
00:37:13.299 --> 00:37:18.049
Myles Brown: And so you could use that for compliance purposes, or auditing, or something like that.

228
00:37:18.519 --> 00:37:26.819
Myles Brown: So they always had kind of a thing that they would do internally to make sure that the data hadn't gotten corrupted. Now you can do it yourself.

229
00:37:30.139 --> 00:37:38.839
Myles Brown: Kind of another thing that is probably worth getting into at some point is the idea of a post-quantum TLS key exchange.

230
00:37:39.189 --> 00:37:44.049
Myles Brown: And so this is sort of an announcement they made, I think this was in November, yeah.

231
00:37:44.199 --> 00:37:53.319
Myles Brown: This concept of post-quantum TLS key exchange, right? There's a very big concern right now, you know, quantum computers

232
00:37:53.319 --> 00:38:04.479
Myles Brown: they're not quite there, right? Like, there's people talking about it, there's… there's companies making money on it already, but it's not really there yet. But we know that eventually we're gonna get to the point where

233
00:38:04.539 --> 00:38:08.699
Myles Brown: Quantum computers can go and, you know.

234
00:38:08.849 --> 00:38:26.269
Myles Brown: be able to break TLS, you know, encryption, even 256-bit encryption, right? And the concern is, right now, people are going to be able to grab that encrypted data, hold onto it, until we have the tools to decrypt it.

235
00:38:26.399 --> 00:38:32.209
Myles Brown: Right? So that's a concern already, like, the concern is here now, even though the tools aren't there yet.

236
00:38:32.859 --> 00:38:45.269
Myles Brown: And so, this idea of post-quantum TLS key encryption is, basically… Let's see… They don't explain it very…

237
00:38:45.829 --> 00:38:55.489
Myles Brown: So the new post-quantum TLS key supports quantum-resistant cryptography for the encryption of data in transit, combined with their server-side encryption.

238
00:38:55.779 --> 00:39:08.119
Myles Brown: Default using 256 algorithms, yeah. So, so AWS is the first one that really kind of jumped in with this idea, and so I think you're gonna see it. They first introduced it with KMS,

239
00:39:08.439 --> 00:39:12.729
Myles Brown: And, the, the, the, what do they call the…

240
00:39:13.749 --> 00:39:17.669
Myles Brown: Secrets Manager? Now, even S3 has it.

241
00:39:18.299 --> 00:39:27.419
Myles Brown: So… all regional S3, S3 tables, Express One Zone endpoints now support MLKEM.

242
00:39:28.249 --> 00:39:35.509
Myles Brown: standards for post-quantum crypto… cryptographic algorithms. So, something to think about now, you know, like I said.

243
00:39:35.619 --> 00:39:42.459
Myles Brown: I… you know, whenever anybody talks about quantum computing, I'm always like, okay, yeah, we're not there yet, right? But…

244
00:39:43.519 --> 00:39:51.469
Myles Brown: the concerns that it's gonna raise are already a problem, right? Because people can grab it, harvest.

245
00:39:51.749 --> 00:39:55.559
Myles Brown: Encrypted data now, and decrypt it at some point in the future.

246
00:39:56.489 --> 00:39:58.079
Myles Brown: So, something to think about.

247
00:39:58.699 --> 00:40:10.959
Myles Brown: Like I said, most of the big concentration of AWS over the last year has been really in AI side of things, and we're not gonna get too much into it, but when it comes to, like.

248
00:40:11.259 --> 00:40:29.129
Myles Brown: you know, data and analytics stack, there are some things that are happening in S3 and open search service, really to get that data ready for AI. And so a big thing is trying to transform S3 from a passive bitstore into a high-performance engine for generative AI.

249
00:40:29.589 --> 00:40:32.419
Myles Brown: And so they basically added a new

250
00:40:33.209 --> 00:40:37.099
Myles Brown: feature to S3 called S3 Vectors. So it's a built-in

251
00:40:37.239 --> 00:40:42.139
Myles Brown: vector search capability for S3, supports up to 2 billion vectors per index.

252
00:40:42.479 --> 00:40:45.639
Myles Brown: So you can basically do RAG, if you know what RAG is.

253
00:40:45.759 --> 00:40:52.189
Myles Brown: Directly on data in your data lake at 90% lower cost than using specialized vector databases.

254
00:40:52.189 --> 00:41:06.489
Myles Brown: Now, specialized vector databases are very popular, and if you're doing that now, you're probably using open search service, and so the big announcement that they made recently with that is that they really improved the vector database performance using GPUs under the covers.

255
00:41:06.919 --> 00:41:13.159
Myles Brown: So, if you've not heard of OpenSearch, that was, AWS went and forked off of Elasticsearch.

256
00:41:13.389 --> 00:41:19.169
Myles Brown: Elasticsearch is an open source… well, was an open source,

257
00:41:19.829 --> 00:41:34.929
Myles Brown: like a NoSQL database, basically. But the company behind it, Elastic, started to really push their agenda a lot, and so AWS said, well, we're gonna fork off of that and make a truly open source version called OpenSearch.

258
00:41:35.099 --> 00:41:38.039
Myles Brown: And then they built their service, open search service, around it.

259
00:41:38.309 --> 00:41:58.039
Myles Brown: Now, usually when I hear, like, a massive company like Amazon is forking off of an open source project, I usually think, oh, they're trying to do something nasty. But here, I think it made a lot of sense. It was Elastic that was pushing their way around more than the community wanted, and so that's what Open Search Service is.

260
00:41:58.639 --> 00:42:00.309
Myles Brown: Alright,

261
00:42:00.459 --> 00:42:11.229
Myles Brown: Oh, I mentioned certification changes, so this is the, certifications that we currently have. At the foundational level, we have the cloud practitioner and the, you know, somewhat newer AI practitioner.

262
00:42:11.529 --> 00:42:24.129
Myles Brown: These are very, very much entry-level, non-technical, really. You know, like, like our salespeople go and get the cloud practitioner, and the AI practitioner is definitely within, you know.

263
00:42:24.389 --> 00:42:34.799
Myles Brown: you don't have to know a lot about AI, you can go in and just look at some free resources, or take a one-day class and do a little bit of study. You're gonna be ready for that.

264
00:42:35.059 --> 00:42:42.719
Myles Brown: Then we've got the associate level, and so the three that have been there forever are the solution architect, the developer, and the cloud ops engineer.

265
00:42:44.019 --> 00:42:48.269
Myles Brown: There is now a Machine Learning Engineer Associate.

266
00:42:48.459 --> 00:42:54.679
Myles Brown: They added that because they removed, or they're about to remove, the machine learning specialty.

267
00:42:55.939 --> 00:42:58.389
Myles Brown: So we now have a machine learning engineer.

268
00:42:58.949 --> 00:43:13.349
Myles Brown: And, we now have the Data Engineer Associate. So when we look at these, the newest one is here, the Generative AI Developer Professional. It was in beta, I think it's still in beta, the beta is about to end, but…

269
00:43:13.349 --> 00:43:19.249
Myles Brown: you know, that's our newest certification. It is a professional level one, right? So,

270
00:43:19.989 --> 00:43:26.929
Myles Brown: you don't just jump from the AI practitioner to that, you should probably get the machine learning engineer first, and just understand that level.

271
00:43:27.569 --> 00:43:40.319
Myles Brown: This machine learning specialty is going away March 31st, so that's gonna be gone. Those other two that are sort of pertinent to the data and analytics stack, the database specialty, it's already gone.

272
00:43:40.519 --> 00:43:42.759
Myles Brown: And the,

273
00:43:43.329 --> 00:43:54.259
Myles Brown: Oh, the data analytics professional is gone. And so instead, we have this newish one, the data engineer associate. That came out a year and a bit ago.

274
00:43:54.589 --> 00:44:12.969
Myles Brown: But the thing that's new about that is we now have a couple of classes that will help you prepare for that data engineer associate. So there's a three-day class that basically covers those same topics, and we have a one-day exam prep class, which is really very much focused

275
00:44:12.969 --> 00:44:17.579
Myles Brown: you know, a lot of times what Amazon does is they come out with, like, an associate-level

276
00:44:17.669 --> 00:44:23.679
Myles Brown: certification, they make a 3-day class that covers a lot of the same topics, probably 70-80%.

277
00:44:23.969 --> 00:44:31.669
Myles Brown: But it's not exam-focused, it's really, like, I need to be a data engineer, right? And then they have the one-day class that is exam prep.

278
00:44:31.839 --> 00:44:34.529
Myles Brown: Right? And so that's kind of how they do things.

279
00:44:34.789 --> 00:44:50.809
Myles Brown: If you are looking to get a class coming up, we have some public classes still. You know, the basic one, if you're new to AWS, is the Architecting on AWS. It's the broadest one, it's the most popular one if you're trying to get the associate.

280
00:44:51.059 --> 00:44:55.249
Myles Brown: solution architect, you know, that's a good start.

281
00:44:55.519 --> 00:45:10.439
Myles Brown: Specifically around, the data analytics stack, we do have the, the full 4-day class called Building Modern Data Analytics Solutions, and we run that sometimes, you know, one in February, one in April.

282
00:45:10.439 --> 00:45:18.529
Myles Brown: It's a four-day class. You can also just buy one of the four days, like, it's actually broken up into four one-day classes as well.

283
00:45:19.609 --> 00:45:33.729
Myles Brown: Where one class is on data lakes, one is on streaming, one is on batch solutions, so EMR. The streaming one covers both Kinesis and managed streams for Kafka, and then a one-day redshift class.

284
00:45:34.679 --> 00:45:40.989
Myles Brown: So that, that's what makes up the four one-day classes that make up that four-day class.

285
00:45:41.329 --> 00:45:55.259
Myles Brown: If you are really interested in storage, we do have a one-day storage, or two-day storage class called Designing and Implementing Storage. It's a lot about S3, a lot about more than just S3. Talks about elastic file system and things like that.

286
00:45:57.119 --> 00:46:13.049
Myles Brown: Building modern applications with AWS NoSQL databases. This is going to cover, of course, DynamoDB, but lots of those other ones, graph databases, things like that. That's a one-day class. And then the data engineering, that three-day class that I mentioned. Our next one is until June.

287
00:46:13.049 --> 00:46:18.819
Myles Brown: Of course, this is just the public schedule. If you're interested in, you know.

288
00:46:19.009 --> 00:46:32.729
Myles Brown: you've got a group of people and you want to get a class going, you can contact us anytime. These are hyperlinks, so I'm going to share this deck. I think there'll be an email that gets sent to everybody who registered for the course, probably tomorrow.

289
00:46:32.879 --> 00:46:42.889
Myles Brown: And so we'll email you this. Oh, yeah, there, Anne just put it in the chat. But yeah, that's the link to our AWS training.

290
00:46:43.239 --> 00:46:44.449
Myles Brown: And so, if you…

291
00:46:45.019 --> 00:46:53.419
Myles Brown: We'll fly down through that. We've got it kind of broken up. If you're really interested in the data and analytics stack, this is the set of courses that make that up.

292
00:46:53.919 --> 00:47:05.299
Myles Brown: So those individual one-day classes, the data lake, the redshift one, the streaming, the batch, and then the full 4-day class, and that's some of the other ones we've got.

293
00:47:06.569 --> 00:47:10.669
Myles Brown: I mentioned that, we do have some new

294
00:47:11.039 --> 00:47:27.869
Myles Brown: AI classes that are just… they're just being added by AWS next week. We just threw them on our website already. But they're all about Agentic AI, because that's sort of the latest area. I imagine AWS… I mean, this is conjecture, but I think they'll probably have an Agentic AI

295
00:47:28.269 --> 00:47:39.929
Myles Brown: certification at some point. They now have 3 one-day classes. There's the Agentic AI Foundations. You can walk into that just off the street, right? There's not a heavy-duty prereq for that.

296
00:47:40.309 --> 00:47:45.469
Myles Brown: And then the next one would be the Building Agentic AI with Amazon Bedrock Agent Core.

297
00:47:45.719 --> 00:47:56.989
Myles Brown: And then the Advanced Generative AI development on AWS. So those are 3 one-day classes. You would probably take them in that order. The Foundations, the agent core, and then the Advanced

298
00:47:57.469 --> 00:48:02.659
Myles Brown: No, not Advanced Generative AI. There, Advanced Agendic Systems. Sorry.

299
00:48:03.329 --> 00:48:08.429
Myles Brown: I haven't added that yet, but we do have,

300
00:48:08.549 --> 00:48:17.459
Myles Brown: And updated… well, updated as of before these classes just got added, but we've got this document that kind of shows

301
00:48:18.149 --> 00:48:19.429
Myles Brown: the,

302
00:48:20.649 --> 00:48:28.569
Myles Brown: The different paths, and, like, if you're interested in certification, and you were to take all the classes that would map to that certification.

303
00:48:28.729 --> 00:48:32.389
Myles Brown: We've got that kind of a thing, so you can get that certification roadmap.

304
00:48:32.799 --> 00:48:39.289
Myles Brown: If you go to, not to the training courses, but to the certifications part, and then scroll down, there's a little button there.

305
00:48:39.719 --> 00:48:45.629
Myles Brown: white button on a white background, but it is a button. But this is where we talk about the certifications.

306
00:48:45.809 --> 00:48:51.589
Myles Brown: So, that's, that's something to look for. Anne, was there anything else we had to talk about?

307
00:48:52.420 --> 00:48:55.739
Anne Fernandez: You… I think you… you… you said it all.

308
00:48:55.740 --> 00:48:56.360
Myles Brown: Hmm.

309
00:48:56.680 --> 00:48:59.170
Anne Fernandez: That was… it was… it was inspiring.

310
00:48:59.200 --> 00:49:05.179
Myles Brown: That's what we said at the beginning, that it was going to be inspired there, there. I don't know how to inspire.

311
00:49:05.180 --> 00:49:18.140
Anne Fernandez: Sorry, guys, if, whoever wasn't here for the very beginning, I was talking about some of the student comments that Miles got for a course that he taught, and somebody said, great job inspiring, even.

312
00:49:18.390 --> 00:49:20.439
Myles Brown: There, I just threw mine.

313
00:49:20.440 --> 00:49:31.100
Anne Fernandez: best comment ever. Yeah. But yeah, so, sometime tomorrow, we'll send out an email, and I think, Miles, did you… we're also going to provide the slides?

314
00:49:31.100 --> 00:49:41.719
Myles Brown: Yes, yeah, I'm gonna provide these slides, because the slides have a bunch of hyperlinks to the, to the announcements and stuff, so I think it'll be good to get the slides, so we'll include that in that email. I put my email in there.

315
00:49:41.750 --> 00:49:55.389
Myles Brown: In the chat, so if you just want to click on that and store it somewhere. If you have questions about the data analytics stack or any AWS courses, you know, like, if you hit me up, I'll put you in touch with, like, the salesperson where, you know, if you're trying to

316
00:49:55.390 --> 00:50:07.329
Myles Brown: put a class together or something. That's not really my area, but I can… I can find the right salesperson for you. If you just go through the website, you know, eventually somebody will contact you for… if you're trying to set up a class.

317
00:50:08.240 --> 00:50:17.100
Myles Brown: We have a few more minutes. If anybody has any questions, you can throw them in the chat, or maybe I can allow people to unmute themselves? Can we do that?

318
00:50:17.760 --> 00:50:20.189
Myles Brown: I don't know if we can do that, let's see.

319
00:50:20.190 --> 00:50:21.100
Anne Fernandez: Sure.

320
00:50:23.370 --> 00:50:31.060
Anne Fernandez: Maybe not through this, let's see, panelists, chat… Yeah, I'm clicking allow to talk.

321
00:50:31.060 --> 00:50:31.630
Myles Brown: Yeah.

322
00:50:31.630 --> 00:50:37.670
Anne Fernandez: Well, actually, I tried it on on Alex, if you're with us, Alex.

323
00:50:37.830 --> 00:50:47.870
Anne Fernandez: It looks like you're, like, muted right now, but perhaps if you were to unmute, you could… you could speak, so if we did allow to talk, you know, you can just raise your hand, and we can try it with you if you would like to.

324
00:50:48.360 --> 00:50:51.870
Anne Fernandez: Say a question, or if you prefer to type it, that's fine too.

325
00:50:53.260 --> 00:50:56.239
Myles Brown: Let me just, stop my share.

326
00:50:58.750 --> 00:51:01.819
Myles Brown: And they even raise their hand, is a good question.

327
00:51:02.360 --> 00:51:05.869
Anne Fernandez: Yeah, there should be a raise hand in the bottom,

328
00:51:06.830 --> 00:51:09.599
Anne Fernandez: With all the different options, perhaps?

329
00:51:09.600 --> 00:51:18.879
Myles Brown: Here, Muhammad asks, can you go over the AI classes in sequence one more time? How far does it get a beginner to be after finishing them? Okay, yeah, so…

330
00:51:19.040 --> 00:51:22.240
Myles Brown: Probably the best way to show this is…

331
00:51:22.890 --> 00:51:25.829
Myles Brown: Just to bring up that certification.

332
00:51:26.990 --> 00:51:28.059
Myles Brown: Kind of thing.

333
00:51:30.040 --> 00:51:42.549
Myles Brown: So, what I would say is this, like, if you're… if you're absolutely new to AWS, you should probably take something like the One Day Tech Essentials, just to be familiar with the console before you go into something like that.

334
00:51:42.980 --> 00:51:49.920
Myles Brown: Hopefully you've got some AWS experience, but, there is…

335
00:51:50.040 --> 00:51:55.090
Myles Brown: a very simple one-day generative AI essentials class that is, you know.

336
00:51:55.380 --> 00:52:13.970
Myles Brown: if you're a newbie, you've never done anything with GenAI, maybe you use ChatGPT a little bit, you know, there is that class, and then there's a prep class that goes with it to get ready for the practitioner. Most people don't need the prep class, like, you know, if you just take that one-day class, you're good to go, probably with a little study.

337
00:52:14.180 --> 00:52:32.079
Myles Brown: And so that's… that's the sort of fundamental. If you're more of a developer who's going to be building agentic systems and things like that, well, then I would suggest you need more AWS background. Maybe the developing or the architecting class, or, you know, if you've been using AWS for a couple years, then you're… you can jump right in.

338
00:52:32.290 --> 00:52:46.909
Myles Brown: And so there's this generative… developing generative AI applications, it's a 2-day class, and then a 3-day advanced class. So that's just on the basic generative AI. It doesn't cover that much of agentic stuff, right? It's just, you know.

339
00:52:47.050 --> 00:52:56.339
Myles Brown: I'm gonna take a large language model and maybe customize it, you know, retrain it with my own data, get it ready.

340
00:52:56.500 --> 00:53:12.989
Myles Brown: put an API on it, and then, you know, use it from some application, right? So, to be able to do that kind of stuff, that's what this two-day, and then maybe even the 3-day class, those will also get you ready for that professional certification, if that's, like, a big goal, right?

341
00:53:13.570 --> 00:53:23.220
Myles Brown: A genteic side of things is a little different. I would say you should probably still take the Gen AI ascent, like, you better have some grounding in generative AI.

342
00:53:23.580 --> 00:53:33.030
Myles Brown: And then after that, there's these three classes. Where'd they go? The first one is Agentic AI Foundations.

343
00:53:33.880 --> 00:53:40.519
Myles Brown: Then, the Agent Core class, And then, the advanced agentic.

344
00:53:40.940 --> 00:53:42.390
Myles Brown: So those are the three.

345
00:53:44.570 --> 00:53:48.280
Myles Brown: I'm gonna redo that, that little…

346
00:53:48.410 --> 00:53:51.440
Myles Brown: learning path document to include an agendic

347
00:53:51.740 --> 00:53:55.169
Myles Brown: We'll get that up there in the next week, for sure.

348
00:53:55.810 --> 00:53:59.089
Anne Fernandez: We always have so many updates to that thing. Thank you for keeping your eye on it.

349
00:53:59.350 --> 00:54:05.170
Myles Brown: Yeah, yeah. So, I think once you get through those 3 classes, like, if you view that.

350
00:54:05.350 --> 00:54:09.070
Myles Brown: Advanced agentic class and what it's covering.

351
00:54:09.380 --> 00:54:11.349
Myles Brown: It's pretty good, right?

352
00:54:12.560 --> 00:54:17.119
Myles Brown: You know, it does expect you to have those two prerequisites, so take those two courses.

353
00:54:17.890 --> 00:54:24.680
Myles Brown: And so it's really about, you know, multi-agent architectures, looking at the different patterns that you would use for that.

354
00:54:24.810 --> 00:54:29.629
Myles Brown: That's a big thing that… That we're finding is that

355
00:54:30.390 --> 00:54:40.869
Myles Brown: you know, the whole point of Magentic AI is that you don't just one… make one big agent that solves all your problems. It's a bunch of little agents that work,

356
00:54:40.920 --> 00:54:52.170
Myles Brown: you know, together, and so there's patterns on how to put those together. And then it also talks about things like, well, we need policies to control access through here.

357
00:54:52.240 --> 00:55:04.150
Myles Brown: We need to, pass the context around, so we need to compress it and optimize it a bit. We need to make sure our network's set up so that these things can talk to one another.

358
00:55:04.880 --> 00:55:08.070
Myles Brown: Really just going beyond the basics of, you know.

359
00:55:08.460 --> 00:55:14.849
Myles Brown: what's an agent, and how do I set up each individual agent to how do I build a real architecture around all this?

360
00:55:15.640 --> 00:55:22.750
Myles Brown: And so that's what they're trying to get you to. But again, to be able to do this in AWS, you better have the AWS knowledge as well.

361
00:55:22.940 --> 00:55:25.850
Myles Brown: So, I think most people here probably do.

362
00:55:26.840 --> 00:55:32.320
Myles Brown: Oh, I was pointing to that and not showing you the… I wasn't sharing my page, was I?

363
00:55:34.030 --> 00:55:35.370
Myles Brown: Yeah, oops.

364
00:55:36.520 --> 00:55:41.149
Myles Brown: I was hand-waving and pointing to something that you weren't seeing. Oops.

365
00:55:41.430 --> 00:55:50.659
Myles Brown: Sorry. Alright, I think… Are there more questions? We've still got a little bit of time.

366
00:55:55.580 --> 00:55:56.490
Myles Brown: I didn't know.

367
00:55:56.490 --> 00:56:00.170
Anne Fernandez: the URL for that FAQ that you've developed.

368
00:56:00.560 --> 00:56:01.150
Myles Brown: Oh, yeah.

369
00:56:01.150 --> 00:56:11.870
Anne Fernandez: For the Gen AI on… on AWS. So there's just, you know, some good questions that Miles answered there, and also, is that the one with a little bit of video in it as well?

370
00:56:11.870 --> 00:56:12.420
Myles Brown: Yeah.

371
00:56:12.620 --> 00:56:13.030
Anne Fernandez: Okay.

372
00:56:13.030 --> 00:56:14.819
Myles Brown: Yeah, and I think that's probably…

373
00:56:15.040 --> 00:56:20.909
Myles Brown: That's probably going to be our next webinar, is sort of, what's the current state of generative AI,

374
00:56:22.130 --> 00:56:23.519
Myles Brown: at AWS.

375
00:56:24.280 --> 00:56:35.770
Myles Brown: Because it is… it's a moving target, like, you know, I could do one of those every 6 months, and they would be, you know, different. But, but I think, you know, we're seeing a lot of…

376
00:56:36.170 --> 00:56:43.400
Myles Brown: you know, agentic AI is what everybody's talking about these days. You know, really moving beyond chatbots. And,

377
00:56:43.780 --> 00:56:47.360
Myles Brown: And so that's… that's the area where we're probably gonna need…

378
00:56:48.210 --> 00:56:58.429
Myles Brown: you know, we'll get a webinar out there, give you an idea of what's the toolset that AWS has, and then, you know, if you want a deeper dive, then you'll go take one of those three classes.

379
00:57:01.480 --> 00:57:02.340
Myles Brown: Alright.

380
00:57:03.590 --> 00:57:07.969
Myles Brown: Well… I think that's all we have for today.

381
00:57:08.970 --> 00:57:12.070
Myles Brown: I'll clean up my slides a little bit, and then,

382
00:57:12.300 --> 00:57:14.579
Myles Brown: Our marketing people will send that out tomorrow.

383
00:57:15.350 --> 00:57:21.439
Anne Fernandez: Yes, we will. So, yeah, we'll give a little swag bag of your webinar slides.

384
00:57:21.620 --> 00:57:35.149
Anne Fernandez: the links of the classes that you went over, maybe even the course, the certification map. I mean, with the caveat we know will be updated, but until now, this video…

385
00:57:35.310 --> 00:57:52.109
Anne Fernandez: links to courses, your email address, we'll just have a chock full of stuff for everyone who attended, and thank you all so much for coming and spending this hour with us. We really appreciate it. Hope you got a lot out of it. I know I did, and thank you, Miles, for your time for this, too. I know you're also very busy.

386
00:57:52.110 --> 00:57:53.980
Myles Brown: Alright, thank you, Ann. This was great.

387
00:57:54.620 --> 00:58:04.039
Anne Fernandez: All right, great. Well, okay, well, we're gonna hold you to the next, the next webinar in 3 to 6 months. You heard it, you heard it, he said it live, everyone.

388
00:58:04.040 --> 00:58:06.680
Myles Brown: Yeah, I was… I said… I said 2 months.

389
00:58:08.300 --> 00:58:08.940
Myles Brown: Oh, you can't hold me.

390
00:58:08.940 --> 00:58:14.000
Anne Fernandez: Okay, then 6 weeks to 2 months, you're on. Very good.

391
00:58:14.580 --> 00:58:28.910
Anne Fernandez: Okay, all right, I guess we can release everyone out into the world now, and we hope to see you next time. We'll let you know for sure next time Miles is up for more AWS webinar content. Okay, thank you, everybody. Bye-bye.

What's New in the AWS Data and Analytics Stack for 2026 [Webinar]

Join Us to Learn About

Audience

About the presenter:

Alert!

Modal Title

Error!

Default Title

Prompt

Confirm

Login