1 00:00:00,120 --> 00:00:01,980 ‫So now let's talk about the differences 2 00:00:01,980 --> 00:00:05,520 ‫of EBS volumes and EFS file systems. 3 00:00:05,520 --> 00:00:07,050 ‫So the EBS volumes, 4 00:00:07,050 --> 00:00:09,450 ‫they're attached to one instance at a time, 5 00:00:09,450 --> 00:00:12,810 ‫except in the edge case of using the multi-attach feature 6 00:00:12,810 --> 00:00:15,690 ‫of the io1 and io2 types of volume, 7 00:00:15,690 --> 00:00:18,480 ‫but that is for very specific use cases. 8 00:00:18,480 --> 00:00:21,960 ‫EBS volumes are also locked at the AZ level. 9 00:00:21,960 --> 00:00:23,490 ‫So here is an example. 10 00:00:23,490 --> 00:00:25,470 ‫We have one ec2 in AZ1, 11 00:00:25,470 --> 00:00:28,050 ‫and we have one EBS volume attached to it, 12 00:00:28,050 --> 00:00:31,083 ‫and it cannot be attached to an ec2 instance in AZ2. 13 00:00:32,040 --> 00:00:34,830 ‫Now for the gp2 type of volumes, 14 00:00:34,830 --> 00:00:37,770 ‫the IO will increase if the disk site increase, 15 00:00:37,770 --> 00:00:41,010 ‫but for io1 you can increase the IO independently. 16 00:00:41,010 --> 00:00:44,220 ‫To migrate an EBS volume across AZ, 17 00:00:44,220 --> 00:00:45,660 ‫we need to take a snapshot 18 00:00:45,660 --> 00:00:48,300 ‫so it will go into the EBS snapshots, 19 00:00:48,300 --> 00:00:51,630 ‫and then we can restore the snapshot into another AZ. 20 00:00:51,630 --> 00:00:54,390 ‫This is how we move from one AZ to the next. 21 00:00:54,390 --> 00:00:58,350 ‫Now, the EBS volumes backups, they will use IO, 22 00:00:58,350 --> 00:00:59,550 ‫and so you shouldn't run them 23 00:00:59,550 --> 00:01:01,830 ‫while your application is handling a lot of traffic 24 00:01:01,830 --> 00:01:04,110 ‫because that may impact the performance. 25 00:01:04,110 --> 00:01:06,120 ‫For your ec2 instances, 26 00:01:06,120 --> 00:01:08,160 ‫the root EBS volumes of your instances 27 00:01:08,160 --> 00:01:10,170 ‫will get terminated by default 28 00:01:10,170 --> 00:01:12,300 ‫if the ec2 instance gets terminated, 29 00:01:12,300 --> 00:01:14,013 ‫but you can disable that behavior. 30 00:01:14,910 --> 00:01:17,040 ‫Now for EFS, it's a bit different. 31 00:01:17,040 --> 00:01:18,660 ‫So it's a network file system 32 00:01:18,660 --> 00:01:20,670 ‫and the goal is really to attach it 33 00:01:20,670 --> 00:01:23,370 ‫to hundreds of instances across availability zones, 34 00:01:23,370 --> 00:01:25,290 ‫so we really see the distinction here. 35 00:01:25,290 --> 00:01:27,750 ‫So with one EFS file system, 36 00:01:27,750 --> 00:01:30,060 ‫we can have different mount targets in different AZs, 37 00:01:30,060 --> 00:01:31,320 ‫and then multiple instances 38 00:01:31,320 --> 00:01:33,900 ‫can share that one file system together. 39 00:01:33,900 --> 00:01:36,150 ‫So it's very helpful, for example, when you have WordPress, 40 00:01:36,150 --> 00:01:38,010 ‫and it's only for Linux instances 41 00:01:38,010 --> 00:01:40,320 ‫because it is using the POXIS system. 42 00:01:40,320 --> 00:01:43,860 ‫The EFS has a higher price point than EBS, 43 00:01:43,860 --> 00:01:48,300 ‫but you can leverage the EFS-IA feature for cost savings. 44 00:01:48,300 --> 00:01:50,130 ‫So hopefully you understand now the difference 45 00:01:50,130 --> 00:01:52,110 ‫between EFS and EBS, 46 00:01:52,110 --> 00:01:53,730 ‫and for the instant store, 47 00:01:53,730 --> 00:01:56,730 ‫well it is physically attached to the ec2 instance 48 00:01:56,730 --> 00:01:59,160 ‫and so therefore if you lose your ec2 instance 49 00:01:59,160 --> 00:02:01,350 ‫you will lose the storage as well. 50 00:02:01,350 --> 00:02:02,580 ‫Alright, that's it. 51 00:02:02,580 --> 00:02:05,703 ‫I hope you liked it and I will see you in the next lecture.