1 00:00:07,450 --> 00:00:09,354 After watching this video, you will be able to: 2 00:00:09,354 --> 00:00:12,658 Recognize how Site Reliability Engineering differs from DevOps 3 00:00:12,658 --> 00:00:16,490 Recognize the commonality between Site Reliability Engineering and DevOps 4 00:00:16,490 --> 00:00:21,570 Explain how Site Reliability Engineering and DevOps can be used together 5 00:00:21,570 --> 00:00:25,829 You might be wondering how DevOps compares with Site Reliability Engineering (SRE). 6 00:00:25,829 --> 00:00:30,650 Before we can explore this topic, we should probably describe what SRE is, how it differs 7 00:00:30,650 --> 00:00:36,020 from DevOps, and how you can leverage SRE in a DevOps environment. 8 00:00:36,020 --> 00:00:40,770 According to Benjamin Treynor Sloss, SRE is “…what happens when a software engineer 9 00:00:40,770 --> 00:00:45,010 is tasked with what used to be called operations.” 10 00:00:45,010 --> 00:00:49,760 Most system administrators are happy doing the same manual tasks day after day perhaps 11 00:00:49,760 --> 00:00:52,760 because they feel that it's their job to do these manual tasks. 12 00:00:52,760 --> 00:00:57,000 But if you ask a software engineer to build a server, they will probably do it manually 13 00:00:57,000 --> 00:00:58,000 the first time. 14 00:00:58,000 --> 00:01:02,830 If a few days later you ask them to build another server just like the first one, they 15 00:01:02,830 --> 00:01:04,260 might do that one manually as well. 16 00:01:04,260 --> 00:01:08,880 But by the time you ask for the third server, a software engineer is going to start writing 17 00:01:08,880 --> 00:01:12,979 a program that builds the server for them automatically. 18 00:01:12,979 --> 00:01:14,819 That's just how software engineers think. 19 00:01:14,819 --> 00:01:18,490 They're programmers; they write programs. 20 00:01:18,490 --> 00:01:23,340 The goal of site reliability engineers is to automate themselves out of a job. 21 00:01:23,340 --> 00:01:28,200 Of course, that will never happen because there are always more things to automate. 22 00:01:29,106 --> 00:01:33,539 One of the tenets of SRE is to only hire software engineers. 23 00:01:33,539 --> 00:01:37,609 You want people who know how to write code so that they can automate repetitive tasks 24 00:01:37,609 --> 00:01:40,270 using Infrastructure as Code. 25 00:01:40,270 --> 00:01:45,259 Site reliability engineers focus on reducing toil, that is, repetitive, manual tasks. 26 00:01:45,259 --> 00:01:49,869 It is recommended that they spend about 50% of their time reducing toil through automation. 27 00:01:49,869 --> 00:01:53,929 The idea is that anything you do repeatedly should be automated. 28 00:01:53,929 --> 00:01:56,729 You shouldn't be doing the same manual task day after day. 29 00:01:57,418 --> 00:02:00,250 SRE teams are separate from development teams. 30 00:02:00,250 --> 00:02:04,039 This is a big difference between DevOps and SRE. 31 00:02:04,039 --> 00:02:08,179 DevOps is the recognition that working in separate siloed teams is inefficient. 32 00:02:08,179 --> 00:02:12,540 But SRE, on the other hand, keeps those silos in place. 33 00:02:12,540 --> 00:02:17,379 The development team is a separate and distinct team from the operations team. 34 00:02:17,379 --> 00:02:21,507 In SRE, stability is controlled through something known as error budgets. 35 00:02:21,507 --> 00:02:25,850 Developers are allowed to deploy their applications into production as long as they don't cause 36 00:02:25,850 --> 00:02:28,349 too many production outages. 37 00:02:28,349 --> 00:02:32,629 The upper limit of allowed outages caused by errors is the error budget. 38 00:02:32,629 --> 00:02:37,189 So, let's say you’ve got a service level agreement of 99.9% uptime. 39 00:02:37,189 --> 00:02:40,069 That equates to about 44 seconds per month of downtime. 40 00:02:40,069 --> 00:02:44,100 As long as the outages are below 44 seconds per month, the developers are free to keep 41 00:02:44,100 --> 00:02:46,709 deploying their releases to production. 42 00:02:46,709 --> 00:02:50,530 Once the developers have caused enough outages to exceed their error budget, they're no longer 43 00:02:50,530 --> 00:02:52,549 allowed to deploy to production. 44 00:02:52,549 --> 00:02:53,680 This actually works pretty well. 45 00:02:53,680 --> 00:02:57,579 It solves the problem of developers waiting for operations, yet it still gives operations 46 00:02:57,579 --> 00:03:02,519 control over the stability of the production environment. 47 00:03:02,519 --> 00:03:07,230 One last thing about SRE is that developers spend about 5% of their time rotating through 48 00:03:07,230 --> 00:03:12,220 the operations team so that they understand what the SRE team is doing on a daily basis. 49 00:03:12,220 --> 00:03:16,530 Also if they cause too many outages, or the toil exceeds 50% of the site reliability engineer’s 50 00:03:16,530 --> 00:03:21,980 time, more developers are shifted to operations to help bring things back into balance. 51 00:03:21,980 --> 00:03:25,969 There is a big difference in teaming between SRE and DevOps. 52 00:03:25,969 --> 00:03:31,060 As we've learned, SRE maintains separate development and operations teams, but it does have one 53 00:03:31,060 --> 00:03:32,879 staffing pool. 54 00:03:32,879 --> 00:03:37,379 That means if you need another site reliability engineer, you take away one of the developers. 55 00:03:37,379 --> 00:03:41,599 If you want another developer, you take away one of the site reliability engineers. 56 00:03:41,599 --> 00:03:43,760 This is an effort to balance things out. 57 00:03:43,760 --> 00:03:49,790 DevOps on the other hand breaks down the silos into one team with one common business objective 58 00:03:49,790 --> 00:03:53,870 to deploy software to production quickly and safely. 59 00:03:53,870 --> 00:03:58,219 The other big difference between DevOps and SRE is how they maintain production stability. 60 00:03:58,219 --> 00:04:04,299 As we said, SRE uses error budgets that development has to comply to and those are based on service-level 61 00:04:04,299 --> 00:04:05,540 objectives. 62 00:04:05,540 --> 00:04:09,670 When a developer exceeds the error budget, making production unstable, they can no longer 63 00:04:09,670 --> 00:04:11,639 deploy to production. 64 00:04:11,639 --> 00:04:18,000 In contrast, DevOps maintains stability by using automation through Continuous Delivery 65 00:04:18,000 --> 00:04:23,720 pipelines, and by making sure that everyone is responsible for the code that runs in production. 66 00:04:23,720 --> 00:04:27,640 DevOps has this “you build it, you run it” mantra. 67 00:04:27,640 --> 00:04:32,480 Unlike SRE, developers are responsible for their applications in production. 68 00:04:32,480 --> 00:04:36,860 There is commonality between the two practices. 69 00:04:36,860 --> 00:04:40,560 Both seek to make development and operations visible to each other. 70 00:04:40,560 --> 00:04:46,660 Whether you have developers rotating through operations as in SRE, or you have development 71 00:04:46,660 --> 00:04:51,880 operations on the same team as in DevOps, everyone understands what it takes to keep 72 00:04:51,880 --> 00:04:54,320 production stable. 73 00:04:54,320 --> 00:04:56,470 Both require a blameless culture. 74 00:04:56,470 --> 00:04:58,980 No one comes to work wanting to take down production. 75 00:04:58,980 --> 00:05:03,280 It’s usually the system that fails the people, not the other way around. 76 00:05:03,280 --> 00:05:07,560 So having a blameless culture is important in both practices. 77 00:05:07,560 --> 00:05:13,510 People can speak openly and honestly about how things are going and how to improve things. 78 00:05:13,510 --> 00:05:17,830 The objective of both is the same—to deploy software faster with stability. 79 00:05:17,830 --> 00:05:24,060 So, DevOps and SRE do have common goals, they just achieve them in completely different 80 00:05:24,060 --> 00:05:26,130 ways. 81 00:05:26,130 --> 00:05:29,910 When we look at how DevOps and SRE can be complement each other and used together, 82 00:05:29,910 --> 00:05:35,470 I like to think of SRE as the team that maintains the infrastructure and DevOps as the team 83 00:05:35,470 --> 00:05:38,430 that uses the infrastructure to maintain their applications. 84 00:05:38,430 --> 00:05:44,580 If you are in a cloud environment, SRE includes the people who operate the cloud and DevOps 85 00:05:44,580 --> 00:05:47,960 includes the people who are consuming the cloud. 86 00:05:47,960 --> 00:05:52,460 This is why using things like platform as a service is so important to DevOps. 87 00:05:52,460 --> 00:05:54,650 The SRE teams provide a platform. 88 00:05:54,650 --> 00:06:00,280 The DevOps teams utilize the platform to deploy their applications. 89 00:06:00,280 --> 00:06:01,860 In this video, you learned that: 90 00:06:01,860 --> 00:06:04,100 SRE takes a different approach than DevOps, 91 00:06:04,687 --> 00:06:06,740 SRE and DevOps have some common goals. 92 00:06:07,620 --> 00:06:12,590 SRE and DevOps can be used together to both maintain and use computer infrastructure.