1
00:00:07,380 --> 00:00:11,930
- The Amazon Athena
service is a great example

2
00:00:11,930 --> 00:00:16,000
of highlighting the
differences between on-prem

3
00:00:16,000 --> 00:00:20,680
infrastructure and cloud
native architecture.

4
00:00:20,680 --> 00:00:21,993
So what is Athena?

5
00:00:22,840 --> 00:00:27,840
It is region scoped and, it's
designed around data analysis

6
00:00:29,170 --> 00:00:33,680
but, it does it in a
special way that really

7
00:00:33,680 --> 00:00:37,800
you can't achieve in
an on-prem environment

8
00:00:38,660 --> 00:00:42,780
because, it utilizes data
that is sitting in S3

9
00:00:43,620 --> 00:00:47,050
but, you can query it as
if it were a relational

10
00:00:47,050 --> 00:00:49,423
database using SQL queries.

11
00:00:50,670 --> 00:00:54,370
The underlying engine
is entirely serverless

12
00:00:55,310 --> 00:00:59,290
but, it utilizes third
party offerings that you

13
00:00:59,290 --> 00:01:03,010
may have heard of, including Presto

14
00:01:03,010 --> 00:01:06,750
for distributing those SQL queries and,

15
00:01:06,750 --> 00:01:11,750
Apache Hive as a metastore
to help understand

16
00:01:12,410 --> 00:01:15,130
how the data is organized

17
00:01:15,130 --> 00:01:18,103
into individual tables and schemas.

18
00:01:20,270 --> 00:01:24,210
Now, in an actual Athena
table, what does it look

19
00:01:24,210 --> 00:01:25,360
like?

20
00:01:25,360 --> 00:01:29,070
Well, the table itself is just a container

21
00:01:29,070 --> 00:01:33,430
a logical resource that
contains the definition

22
00:01:33,430 --> 00:01:34,623
of the metadata.

23
00:01:36,740 --> 00:01:38,839
It specifies the location of the data

24
00:01:38,839 --> 00:01:42,100
in terms of S3 buckets and

25
00:01:42,100 --> 00:01:47,100
prefixes as well as the actual
structure of the data itself.

26
00:01:50,330 --> 00:01:54,050
The tables in Athena don't
actually contain the data.

27
00:01:54,050 --> 00:01:59,050
All of the data is in the
individual objects in S3.

28
00:02:01,390 --> 00:02:04,950
You can create Athena tables automatically

29
00:02:04,950 --> 00:02:07,410
or, you can create them
by hand, especially

30
00:02:07,410 --> 00:02:09,890
if you have a number of disparate sources

31
00:02:09,890 --> 00:02:13,453
for that data that might even
be in other AWS accounts.

32
00:02:14,980 --> 00:02:18,620
And, finally, you can
organize your individual

33
00:02:18,620 --> 00:02:23,620
Athena tables together
to create a database,

34
00:02:24,180 --> 00:02:26,530
as a logical grouping of those tables

35
00:02:26,530 --> 00:02:28,650
if you want to be able to issue queries

36
00:02:28,650 --> 00:02:31,423
across multiple tables at the same time.

37
00:02:33,769 --> 00:02:36,500
Now, the third and most important element

38
00:02:36,500 --> 00:02:39,550
of Athena is called the data catalog.

39
00:02:39,550 --> 00:02:41,810
And, this is really important

40
00:02:41,810 --> 00:02:44,963
for big data and data lake uses.

41
00:02:45,800 --> 00:02:48,740
This is a system for organizing all

42
00:02:48,740 --> 00:02:52,940
of the individual tables
into a cohesive unit

43
00:02:54,120 --> 00:02:59,120
and, the data catalog is what
helps to combine the dataset

44
00:02:59,800 --> 00:03:04,030
which is the actual
underlying data, as well

45
00:03:04,030 --> 00:03:08,210
as the tables and
schemas those definitions

46
00:03:08,210 --> 00:03:11,660
into a data source that
can then be treated

47
00:03:11,660 --> 00:03:13,123
as a logical object.

48
00:03:15,390 --> 00:03:19,520
There's a legacy version of
data catalogs that are managed

49
00:03:19,520 --> 00:03:21,570
by the Athena service itself.

50
00:03:21,570 --> 00:03:22,930
And, these are being phased out.

51
00:03:22,930 --> 00:03:26,723
You may or may not have access
to these in your AWS account.

52
00:03:27,990 --> 00:03:32,230
If you create a data catalog today though,

53
00:03:32,230 --> 00:03:34,940
it actually uses another data analytic

54
00:03:34,940 --> 00:03:39,320
service called 'Glue' that
we're going to be talking about

55
00:03:39,320 --> 00:03:43,723
in a later lesson to help
organize those data catalogs.