1
00:00:00,360 --> 00:00:02,969
So now let's do a summary of Amazon DynamoDB

2
00:00:02,969 --> 00:00:05,100
which appears quite a bit at the exam.

3
00:00:05,100 --> 00:00:09,270
So DynamoDB is a proprietary technology from AWS.

4
00:00:09,270 --> 00:00:12,060
It's a managed serverless NoSQL database

5
00:00:12,060 --> 00:00:15,780
that provides you, out of the box, millisecond latency.

6
00:00:15,780 --> 00:00:18,240
You have two capacity modes you need to choose from:

7
00:00:18,240 --> 00:00:21,240
so one of them is the provisioned capacity

8
00:00:21,240 --> 00:00:22,560
with optional auto-scaling,

9
00:00:22,560 --> 00:00:23,400
which is great

10
00:00:23,400 --> 00:00:25,080
when you have a smooth type of workload

11
00:00:25,080 --> 00:00:26,820
that increases gradually over time,

12
00:00:26,820 --> 00:00:29,130
or decreases gradually over time,

13
00:00:29,130 --> 00:00:31,530
or you have the on-demand capacity mode

14
00:00:31,530 --> 00:00:34,020
where you don't have to provision capacity

15
00:00:34,020 --> 00:00:36,510
but it scales automatically for you, which is great

16
00:00:36,510 --> 00:00:39,210
if you have very unpredictable workloads,

17
00:00:39,210 --> 00:00:41,850
or if you have sudden steep spikes

18
00:00:41,850 --> 00:00:44,370
of demand on your database.

19
00:00:44,370 --> 00:00:47,850
So DynamoDB can replace ElastiCache as a key/value store,

20
00:00:47,850 --> 00:00:50,130
and it's a great way to, for example,

21
00:00:50,130 --> 00:00:52,320
store session data for your website,

22
00:00:52,320 --> 00:00:54,120
combined with a TTL feature

23
00:00:54,120 --> 00:00:58,020
to expire a row after a specific amount of time.

24
00:00:58,020 --> 00:01:01,230
So DynamoDB is highly available,

25
00:01:01,230 --> 00:01:04,769
it is across multiple availability zones by default,

26
00:01:04,769 --> 00:01:07,110
the Reads and the Writes are fully decoupled,

27
00:01:07,110 --> 00:01:09,780
and it's possible for you to have transactions

28
00:01:09,780 --> 00:01:11,403
on top of DynamoDB tables.

29
00:01:12,300 --> 00:01:14,730
It is possible for you to create a read cache

30
00:01:14,730 --> 00:01:17,580
out of the box that is fully compatible with DynamoDB,

31
00:01:17,580 --> 00:01:21,570
which is a DAX cluster DynamoDB accelerator

32
00:01:21,570 --> 00:01:24,330
and the ability... the particularity of it

33
00:01:24,330 --> 00:01:28,320
is that you get then microsecond read latency.

34
00:01:28,320 --> 00:01:30,420
So look for that in the exam.

35
00:01:30,420 --> 00:01:33,360
In terms of security, authentication, and authorization,

36
00:01:33,360 --> 00:01:36,000
everything is done through IAM.

37
00:01:36,000 --> 00:01:39,540
You have event processing capability on top of DynamoDB,

38
00:01:39,540 --> 00:01:41,760
so you can enable DynamoDB streams

39
00:01:41,760 --> 00:01:45,300
to stream all the changes happening into database,

40
00:01:45,300 --> 00:01:48,840
and you can have an integration to invoke Lambda

41
00:01:48,840 --> 00:01:50,370
from the DynamoDB stream.

42
00:01:50,370 --> 00:01:54,090
Therefore, Lambda can be invoked for every single change

43
00:01:54,090 --> 00:01:56,040
in your DynamoDB table.

44
00:01:56,040 --> 00:01:58,710
The other option is instead of sending the data

45
00:01:58,710 --> 00:02:00,030
to DynamoDB streams,

46
00:02:00,030 --> 00:02:02,820
you can send the data to Kinesis Data Streams.

47
00:02:02,820 --> 00:02:05,190
The advantage is that on top of it, you can use

48
00:02:05,190 --> 00:02:07,410
the Kinesis Data Firehose feature,

49
00:02:07,410 --> 00:02:09,419
or you can use any type of integration

50
00:02:09,419 --> 00:02:11,310
that leverages Kinesis Data Stream.

51
00:02:11,310 --> 00:02:13,350
For example, you could have longer term retention

52
00:02:13,350 --> 00:02:16,740
up to one year on Kinesis Data Streams.

53
00:02:16,740 --> 00:02:19,650
You have a Global Table feature on top of DynamoDB,

54
00:02:19,650 --> 00:02:23,160
which allows you to do active-active replication

55
00:02:23,160 --> 00:02:25,170
across multiple regions.

56
00:02:25,170 --> 00:02:28,140
So anyone can read and write from any region,

57
00:02:28,140 --> 00:02:30,450
this is why it's called active-active.

58
00:02:30,450 --> 00:02:31,800
Now you have two backup options,

59
00:02:31,800 --> 00:02:34,380
the first one is an automated backup.

60
00:02:34,380 --> 00:02:37,050
You need to enable point in time recovery for that,

61
00:02:37,050 --> 00:02:42,050
and you can restore your table to a new DynamoDB table

62
00:02:42,150 --> 00:02:45,390
for any point of time, up to 35 days.

63
00:02:45,390 --> 00:02:49,140
Or if you want to have longer term retention of backups,

64
00:02:49,140 --> 00:02:51,390
you can enable on-demand backups,

65
00:02:51,390 --> 00:02:54,510
which will also restore to a new table.

66
00:02:54,510 --> 00:02:58,200
Now you can export your DynamoDB table to Amazon S3

67
00:02:58,200 --> 00:03:01,200
without using any of the Read Capacity Units

68
00:03:01,200 --> 00:03:03,270
within the point in time recovery window,

69
00:03:03,270 --> 00:03:07,140
so within 35 days using the Export to S3 feature,

70
00:03:07,140 --> 00:03:08,940
and you can also import from S3

71
00:03:08,940 --> 00:03:12,333
without using any Right Capacity Units into a new table.

72
00:03:13,560 --> 00:03:15,600
So DynamoDB from an exam perspective,

73
00:03:15,600 --> 00:03:18,240
anytime you see that you need to rapidly evolve schemas,

74
00:03:18,240 --> 00:03:21,180
you need to have a flexible type of database schema,

75
00:03:21,180 --> 00:03:23,010
it's going to be a great choice.

76
00:03:23,010 --> 00:03:24,962
So the use cases for DynamoDB is:

77
00:03:24,962 --> 00:03:27,210
a serverless application development

78
00:03:27,210 --> 00:03:30,510
for small documents that are less than 400 kilobytes,

79
00:03:30,510 --> 00:03:32,280
distributed serverless cache,

80
00:03:32,280 --> 00:03:33,210
and just so you know,

81
00:03:33,210 --> 00:03:37,590
DynamoDB does not have any SQL query language available.

82
00:03:37,590 --> 00:03:38,520
All right, that's it.

83
00:03:38,520 --> 00:03:41,520
I hope you liked it, and I will see you in the next lecture.