I am a big fan of O'Reilly's 97 Things series and happen to be addicted to that series.?
Thanks and more love to Nathen & Emily for the 97 Things Every Cloud Engineer Should Know, which is collective wisdom from the industry experts for the cloud community.?All credit goes to the authors who initially thought about it; it just my notes and the outcome of the book read.?
- Why The Cloud??Instead of what cloud is, I love the answer and insights from why the cloud? The cloud brings more value when we invest in learning how to automate and compress that cycle. One thing to be noted is there is always room for improvement. Getting Started is always greater than Getting Finished, so we must take an iterative approach. So, don't forget that getting started with improvements is more important than getting finished.?
- Serverless Bad Practices.?Serverless is super and supreme if we architect well with the system's need rather than to pick like a new toy pattern. A few bad practices like calling a function synchronously, calling a function asynchronously, employing many libraries, using more tools, and not documenting functions.?
- It's okay if you're not running Kubernetes:?Because the media tells us half the truth only. There is still much to improve on the old and dull side of the technology, but don't sit still.?
- Why the lift and shift fails:?The cloud service provider and on-prem environment are very different, primarily when operating workloads in a cloud provider environment; the way manage things are pretty other. So, moving from on=prem to cloud takes a lot of thought, effort, coordination, and considerations, so take time to things right.?
- ?You Can't Get Information Security Right Without Getting Identify Right:?Security is Job Zero for every cloud engineer. It is critical to remember that you are building for multiple tenants who should never access or change one another's data in a cloud environment. Every permission has to be perfect every time. Whatever identity experience you build, remember that it's the front door to your application. It needs to be both welcoming and secure.?
- Revisiting the R's of SREs:?One of the hottest topics in site reliability engineering right now is how to make our applications and services resilient in the face of failure. In resilience engineering, those R-words refer to specific and different aspects of the socio-technical systems within which we exist, and maneuver is robustness, reliability, rebound, resilience.?
- There are no such things as a development environment:?Because it is a candidate production system, 1st up, a candidate production system is not the same as a natural live production system and not even closer. Another point is that?the production system?means the system that someone else, anyone else, can hold you accountable. There's more to it than that, though. If you're making your development systems as production-like as practicable (and you are, right? Right?), significant opportunities are available to build, test, and refine the production procedures as early as possible.?
- Incident Analysis and Chaos Engineering are Complementary Practices:?Well-done incident analysis takes time. Constraining from a few hours or filling out a template to complete an investigation does not result in high-quality research—software designed to express specific characteristics and behave in expected ways. Eventually, things will change enough that our initial designs may no longer serve us.??
- KISS it:??Most of us have probably heard the famous acronym KISS, or keep it simple, stupid. Whether you realize it or not, we apply this fantastic guiding principle in our life. It's such a natural idea, but we seem to have the opposite urge regarding our work in IT. Why do we find it so challenging to KISS in IT??
- Do more with less:?Cloud infrastructure introduces many benefits, such as elastic scaling, immutable deployments, and a pay-as-you-go pricing model. A better way of designing our applications to achieve better performance with fewer resources is called the?reactive programming model.?
- Your Greatest Products Are Not the Applications and Services You Produce:?The most excellent product you can deliver to our customers is the superpower. The power-up. The Magic Moment! These are found inside your applications and services, at the intersection between functionality, aesthetics, and surprise and delight. It's when an end-user suddenly encounters an unexpected flash "wow-that-is-really-f***ing-cool" moment, leaving a lasting impression, coloring the way they express themselves as they share this gift with their friends, family, and coworkers.?
- Beyond the Portal: Manage Your Cloud with the CLI:??With the cloud, you can log in to an administration portal with a friendly and intuitive graphical user interface (GUI), and with few mouse clicks and configuration settings, your server is ready. In less than five minutes, you can remotely log in to your newly created server. Meanwhile, give the CLI a try. Understand the command structure, practice, and get comfortable with it. You won't regret it scaling and enterprise-level cloud deployment and development.
- Networking First: Far?too often, organizations forget that the cloud, for all its serverless, PaaSified glory, still relies on actual network routing to get the job done, and we, as a cloud engineer, need to understand the basics to be successful.
- Cloud Engineering Is About Culture, Not Containers:?If integration is complicated, continuous deployment is even more complex. Achieving a high release cadence requires technical skill and hospitable culture. If failure is harshly punished, a team will feel compelled to invest in testing and validation before doing a release. That makes releases too expensive to do often. Even in groups with a cultural safety net, releasing regularly requires nearly 100% automation and sophisticated rollback or roll-forward mechanisms.
- Stay Curious:?As a cloud engineer, you should know a lot. And the more you learn, the more you realize that the learning has no end. It becomes a lifelong journey, with new technologies, platforms, and solutions released almost daily!
- Empathy As Code:??IaC then turns into a form of?heart as code, through which groups of folks such as developers, security engineers, and cloud engineers can find common ground, understand incentives and goals, test new designs, and share responsibility for the system as a whole.
