How the Cloud Made Computing Harder, Not Easier

Remember when cloud computing was supposed to make life easier for IT professionals? PaaS and IaaS relieve IT teams of the burden of managing physical infrastructure SaaS Can you easily deliver applications to any user, anywhere, anytime?

Clearly the cloud has done that. But overall, the cloud IT Engineer (and developers, security teams, pretty much anyone trying their hand at modern IT) harder, no easier.

The questions that need to be answered are: why? Why does cloud computing end up making IT departments’ jobs harder and more time-consuming while exposing businesses to new types of cybersecurity threats? Increased overall IT spending?

Let’s examine this question as we look at how the cloud has evolved and what has resulted in a cloud computing ecosystem far more complex than ever necessary.

Why cloud management is difficult

Again, it’s true that the cloud has simplified computing in certain ways. Most businesses can say goodbye to managing their physical infrastructure. It has also deployed sophisticated tools and services such as Managed. Kubernetes It puts environmental and cloud-based big data services in the hands of IT organizations that struggle to implement these technologies on their own due to staffing or expertise limitations.

But in other ways, the cloud has added a lot of challenges to the lives of IT practitioners.

  • Additional tools to master: Instead of becoming an expert in server administration or learning the ins and outs of Linux CLI tools, today’s IT professionals must master cloud computing tools and services from a variety of vendors.
  • A new type of tooling: The cloud has introduced (or popularized) a specific category of tools that do not exist in an on-premises environment and therefore do not need to be managed. Before the cloud, few IT teams had to worry about, for example, writing IAM policies or figuring out how to connect remote cloud-based workloads to a VPN.
  • A more complex budget: The cloud has allowed enterprises to move to an OpEx model for their infrastructure needs. However, the Byzantine approach makes it much more difficult to reliably estimate costs in many cases. pricing schedule It is something the cloud provider imposes on the user.
  • Additional options: Cloud computing gives IT teams much broader options for how and where they host their workloads. But it also means they need to spend more time evaluating and validating different approaches. In an on-premises environment, there were limited ways to do things and less time was spent optimizing a strategy.
  • Limited Visibility: In some cases, the amount of visibility and control IT teams have over their cloud workloads is limited. For example, many cloud services only expose certain types of metrics and logs, and rarely allow customers to see what’s happening on the underlying infrastructure. In other words, IT teams are expected to maintain the same (or better) levels of performance and availability as on-prem was at their best, but with less information to operate on.

How to simplify cloud computing

Clearly, much of the cloud’s complexity comes from the flexibility and performance that cloud computing provides. complexity This is the price your IT team pays to take advantage of the cloud.

That said, managing the cloud didn’t have to end up being so difficult. And it doesn’t have to remain that way. Here are some ways cloud computing can be made simpler from the perspective of engineers managing cloud workloads.

Standardization across clouds

For starters, cloud providers can do a better job of standardizing how their services are configured. All major clouds offer the same set of core services such as VM hosting, object storage, and IAM. However, the configuration tools and policy frameworks used by each cloud are different enough that it is impossible for an engineer who has mastered one cloud to manage another with the same competence.

This issue will be resolved once the cloud vendors agree on some form of underlying standard. You don’t have to provide the exact same service. That obviously wouldn’t be advantageous from a business point of view. But they can do things like standardize the way IAM policies are written, which will help greatly reduce the complexity of managing the cloud.

A simpler pricing schedule

Simplifying pricing schedules is another way public clouds can significantly ease IT teams’ lives. Instead of considering a complex set of variables when calculating the cost of cloud services, the cloud can do things like offer tiered pricing options where customers pay a consistent and predictable price for a fixed level of consumption. That is, you get a preset level of runtime, egress traffic, API calls, etc., and your workload can consume up to that preset level at a predetermined price.

Obviously, tiered pricing isn’t right for all cloud services, and some customers benefit from paying only for exactly what they use. But offering simpler pricing terms, at least as an option, can help many customers solve the pricing problem of cloud computing.

custom metric

It is unreasonable (not to mention highly insecure) to expect cloud vendors to give their customers full access to the underlying cloud infrastructure. However, vendors can at least provide a way for customers to define custom metrics or pull data from lower levels of the hosting stack within limits.

For example, instead of just being able to collect the metrics that the cloud chooses to expose for the VM hosting service, the customer can run it through the cloud. eBPF A program on the server hosting the VM to collect low-level data through the host kernel. This will open up whole new possibilities for monitoring and observability, as IT engineers will not be limited to the metrics provided by cloud services and the data available inside each VM instance. Instead they can see what the hypervisor and host kernel see.

It’s no big deal from an engineering point of view for customers to do this in a secure way without collecting data about other customers’ workloads. However, it is cloud providers like Amazon and Microsoft who have the development resources to build these types of solutions.

Prefer open source tools

One of the reasons learning how to manage cloud services is difficult is that cloud vendors have a habit of creating proprietary management tools, even when there are good open source alternatives available.

For example, most cloud-based managed Kubernetes services come with various vendor-specific CLI tools such as eksctl and gkectl. These tools are not required because vendors can choose to use open-source Kubernetes tools such as kubectl for this purpose instead. You may need to add proprietary extensions to your tools, but the core tools can be open source.

Conclusion: The cloud is great, but it could be simpler.

To be clear, I am not suggesting that IT teams should abandon the cloud or that cloud vendors are intentionally making their cloud services more difficult to manage than necessary. Instead, the problem is that cloud services and management tools have evolved in ways that make cloud management more complex than necessary.

The good news is that there are actionable ways to reduce (if not completely solve) the cloud complexity problem. Most require some effort on the part of the cloud provider, but there’s no reason it can’t be managed.

About the author

Christopher Torzzi HeadshotChristopher Tozzi is a technology analyst with expertise in cloud computing, application development, open source software, virtualization, containers, and more. He also lectures at major universities in the Albany, New York area. His book “For Fun and Profit: A History of the Free and Open Source Software Revolution” was published by MIT Press.

Source

Also Read :  Aetina Introduces End-to-End AI Management Solution Powered by NVIDIA AI at GTC

Leave a Reply

Your email address will not be published.