1 00:00:00,330 --> 00:00:02,820 So now let's talk about Amazon Timestream 2 00:00:02,820 --> 00:00:04,140 and the name indicates 3 00:00:04,140 --> 00:00:06,630 that it's actually a time series database. 4 00:00:06,630 --> 00:00:08,640 So it's fully managed, it's fast, it's scalable 5 00:00:08,640 --> 00:00:10,320 and it's serverless. 6 00:00:10,320 --> 00:00:12,000 So what is a time series? 7 00:00:12,000 --> 00:00:15,240 Well, it's a bunch of points that have a time 8 00:00:15,240 --> 00:00:16,079 included in them. 9 00:00:16,079 --> 00:00:18,750 So for example, here's a graph by year. 10 00:00:18,750 --> 00:00:20,340 So this is a time series. 11 00:00:20,340 --> 00:00:22,440 Now with Timestream you can automatically 12 00:00:22,440 --> 00:00:26,670 adjust the database up and down to scale capacity 13 00:00:26,670 --> 00:00:30,150 and you can store and analyze trillions of events per day. 14 00:00:30,150 --> 00:00:33,150 The idea is that if you have a time series database 15 00:00:33,150 --> 00:00:36,600 it's going to be much faster and much cheaper 16 00:00:36,600 --> 00:00:40,290 than using relational databases for time series data. 17 00:00:40,290 --> 00:00:44,190 Hence the specificity of having a time series database. 18 00:00:44,190 --> 00:00:45,420 You can do schedule queries. 19 00:00:45,420 --> 00:00:47,940 You can have records with multiple measures 20 00:00:47,940 --> 00:00:50,880 and there is full SQL compatibility. 21 00:00:50,880 --> 00:00:53,490 The recent data will be kept in memory. 22 00:00:53,490 --> 00:00:55,470 And then the historical data is kept 23 00:00:55,470 --> 00:00:58,350 in a cost-optimized storage tier. 24 00:00:58,350 --> 00:01:01,620 As well as, you have time series analytics function 25 00:01:01,620 --> 00:01:04,349 to help you analyze your data and find patterns 26 00:01:04,349 --> 00:01:05,940 in near real time. 27 00:01:05,940 --> 00:01:08,850 This database just like every database on AWS 28 00:01:08,850 --> 00:01:11,640 supports encryption in transit and at rest. 29 00:01:11,640 --> 00:01:14,640 So the use cases for Timestream would be 30 00:01:14,640 --> 00:01:16,350 to have an IoT application, 31 00:01:16,350 --> 00:01:18,720 operational applications, real-time analytics 32 00:01:18,720 --> 00:01:22,470 but everything related to a time series database. 33 00:01:22,470 --> 00:01:25,500 Now in terms of architecture, Timestream is here 34 00:01:25,500 --> 00:01:29,460 and it can receive data from AWS IoT, so internet of things. 35 00:01:29,460 --> 00:01:32,250 Kinesis Data Streams through Lambda can receive data 36 00:01:32,250 --> 00:01:33,083 as well. 37 00:01:33,083 --> 00:01:36,150 Prometheus, Telegraf, there are integrations for that. 38 00:01:36,150 --> 00:01:38,160 Kinesis Data Streams as well through 39 00:01:38,160 --> 00:01:40,380 Kinesis Data Analytics for Apache Flink 40 00:01:40,380 --> 00:01:43,500 can receive data into Amazon Timestream 41 00:01:43,500 --> 00:01:46,560 and Amazon MSK as well through the same process. 42 00:01:46,560 --> 00:01:49,290 And in terms of what can connect to Timestream, 43 00:01:49,290 --> 00:01:52,470 where we can build dashboards using Amazon QuickSight. 44 00:01:52,470 --> 00:01:55,230 We can do machine learning using Amazon SageMaker. 45 00:01:55,230 --> 00:01:57,900 We can do Grafana or because there is a standard 46 00:01:57,900 --> 00:02:00,150 JDBC connection into your database, 47 00:02:00,150 --> 00:02:02,580 any application that is compatible with JDBC 48 00:02:02,580 --> 00:02:06,060 and SQL can leverage Amazon Timestream. 49 00:02:06,060 --> 00:02:06,893 So that's it. 50 00:02:06,893 --> 00:02:08,520 I think for the exam you just need to remember 51 00:02:08,520 --> 00:02:10,080 what Timestream is at a high level 52 00:02:10,080 --> 00:02:12,630 but I want to give you a bit more details as well. 53 00:02:12,630 --> 00:02:14,010 So that's it for this lecture. 54 00:02:14,010 --> 00:02:14,850 I hope you liked it 55 00:02:14,850 --> 00:02:16,863 and I will see you in the next lecture.