Formed in 2009, the Archive Team (not to be confused with the archive.org Archive-It Team) is a rogue archivist collective dedicated to saving copies of rapidly dying or deleted websites for the sake of history and digital heritage. The group is 100% composed of volunteers and interested parties, and has expanded into a large amount of related projects for saving online and digital history.
Although a relatively new concept, site reliability engineers (SREs) have become crucial for DevOps teams, helping to solve an array of operational problems such as network availability and user experience. However, in previous years, some people have questioned the longevity of such a role.Â
This article will discuss several reasons why site reliability engineering roles are here to stay and why they have become critical to DevOps.Â
What Are Site Reliability Engineers (SREs)?
A site reliability engineer helps connect development teams and IT operations by completing tasks previously assigned to operational roles. SREs use a range of automation tools for troubleshooting purposes, creating software systems that are both scalable and reliable.
The role revolves around automation and standardization, as cloud systems become increasingly prevalent. SREs also play a vital role in ensuring the user experience of systems is of the highest standard.
Why Site Reliability Engineering Roles Are Here to Stay
Let’s take a look at two of the primary reasons SRE roles are here to stay.Â
User Expectations
The key reason for the rise in popularity and the expected longevity of site reliability engineering roles is user experience. In the modern climate, user expectations are very high and more dedicated roles are required to help deliver a service that is not only acceptable but attempts to exceed user expectations.
Even back in the 2010s, users were much more patient in terms of website load speeds and how quickly they could access information. Even crashes and load failures were tolerated without the user exiting a website.
However, that has all changed due to the competitiveness of the market, as well as the growth of remote working, which has increased dependency on online resources. This means that if a service is unavailable for any period, users are unable to do their job.Â
An SRE’s role includes optimizing systems, ensuring they perform at the required level and that resources are always available to those that need them. This makes SREs critical to DevOps teams that may not have the capacity to focus on meeting user expectations sufficiently.Â
The Complexity of Software Architectures
Another factor that has added a lot of value to the role of an SRE is their ability to provide much-needed support for software environments and architectures that continue to develop in terms of complexity. Kubernetes is one such architecture that requires dedicated attention.Â
Applications that are distributed across systems such as Kubernetes and other cloud-native platforms require specialized personnel that have the necessary knowledge and experience. Machine learning is also becoming more prevalent and requires special focus from a performance perspective. In 2020, just more than 22% of companies had machine learning models in production for one to two years.
This level of complexity is only expected to grow further over the coming years as cloud-native systems become the norm. This is why new roles must be created that go beyond that of a normal IT engineer.Â
Challenges for SRE Roles
Like any IT role, the role of a site reliability engineer always faces challenges that have led people to question whether SREs are a long-term solution or whether it is a niche specialization.Â
A Lack Of Opportunity Away From Hyperscale Companies
The role of an SRE was created by hyperscale companies like Google that manage extremely large IT systems. For companies that do not have such a wide and varied IT infrastructure on a global scale, there are questions about whether such a role is necessary.Â
However, as previously mentioned, the growth of cloud-native technology shows no sign of slowing and, with this growth, opportunities for SREs at smaller organizations may arise.Â
The Role of an SRE is Considered Obscure
The ambiguity of an SRE role can be considered a downfall in some cases, with many businesses unsure of what these professionals offer when compared to a traditional systems engineer. In many cases, SREs are deemed to be a hybrid of a software engineer and an IT operations role.Â
Fortunately, DevOps has realized the potential of SREs and their ability to provide an important bridge between developers and engineers, helping to achieve dedicated tasks to a high standard.Â
SREs Vs. DevOps
Although both roles are connected and share some similarities, there are some clear differences.
- Processes – DevOps has full visibility into the development environment which allows them to make changes from the initial development stages through to production. SREs, on the other hand, only have visibility into the production process, allowing them to make suggestions to ensure performance levels are maintained.
- Implementation – The task of implementing new features on a system falls on the DevOps team. Meanwhile, SREs are tasked with ensuring the new features do not cause any system failures or impact performance during the production stage.
- Key Focus – The primary focus of a site reliability engineer is to ensure the system’s reliability, availability and scalability. DevOps focuses on how quickly product development is completed, as well as its continuity.
- Structure – The structure of an SRE team and a DevOps team have obvious differences. SRE teams are made up of individuals who have similar skill sets relating to development and operations. Whereas DevOps teams consist of professionals who have specific roles related to individual aspects of the project. These roles can include the team leader, product owner, cloud architect, software developer, QA engineer, system administrator, release manager and more.
- Tools – DevOps teams often use tools such as integrated development environments (IDEs) when developing a product. These can include Jenkins (continuous integration and continuous development), JIRA (change management), Splunk (log monitoring) and GitHub (distributed version control). SREs commonly use tools such as Prometheus and Grafana (collection and visualization of metrics, such as CPU usage and available disk space), OP5 and PagerDuty (incident alerts), Ansible and Kubernetes (container orchestration) as well as a range of cloud platforms.
- Bug Reporting – DevOps debug code whenever a bug is reported in the end product. SREs are not involved with debugging from a development perspective and are only required to perform such tasks if there is a production outage, there are infrastructure issues or perhaps if they are dealing with common AWS misconfigurations.
- Measuring Performance – The typical metrics that DevOps uses to measure performance are deployment frequency and deployment failure rate. SREs measure service level objectives (SLOs), service level indicators (SLIs), service level agreements (SLAs) and error budgets.
- Incident Handling – In the case of incident handling, DevOps teams work to mitigate an issue based on the feedback provided. SREs then analyze the issues post-incident, including the root cause of the issue. This is then documented and provided to the developers to fix.Â
Why SRE Roles Are Critical to DevOps – Conclusion
As you can see, the two roles differ significantly but SREs can play a critical role in ensuring the DevOps team delivers a product that offers maximum performance.Â
The core benefits of employing professionals in a site reliability engineer capacity are high levels of product performance and reliability to meet user expectations while helping to manage complex IT architecture.Â
In the future, organizations that manage a large amount of cloud-based IT assets may struggle to deliver efficient products without such professionals. As such, demand for SRE roles is likely to grow.Â
