Core Cloud Concepts: Part 4 — Reliability
For the past few weeks, I’ve discussed some of the core concepts of cloud computing. This series was designed for tech professionals (whether at the staff or senior-level management/executive level) in evaluating the potential benefits of a cloud solution.
In the first post, we covered elasticity and the availability of resources. In the second post, we talked about the importance of the isolation of cloud resources. Our third post discussed trusting cloud platforms. The final part of this series will cover cloud reliability.
After all, there’s frankly no need to discuss other concepts or possibilities with cloud computing if the platform has reliability issues. The ongoing pandemic forced plenty of companies to embrace cloud computing to adapt to the pandemic. This creates concern regarding the security and reliability of these systems if they were not properly configured and were not readily available. More than ever, we need to understand the core aspects of cloud reliability to use these platforms most effectively.
I like the way Lucidchart lays out some basic expectations of cloud platforms regarding reliability: they should minimize downtime; data stored on the cloud needs to be accessible from any necessary devices, and the connection to the cloud should be secure. But two ideas we need to keep in mind are that the cloud is not a flawless solution and that customers have a decisive role to play in developing a more robust cloud system.
The General Reliability of Cloud
We first need to understand that no cloud system will flawlessly work one hundred percent of the time. In fact, if you look over your cloud service provider’s service level agreement, they’ll likely write that uptime at or about 99.999999995% vs. 100%. This isn’t meant to be hyperbole; rather, an open-eyed acknowledgment that cloud computing has the inherent risk of downtime. Acts of God, like an earthquake or hurricane, can take out a server. Effective cloud providers embrace distributed cloud to have servers located as close as possible to customers and provide backups in the event of an issue with a specific server. Because of this distribution, the question becomes how quickly service can be restored.
Customers, Providers, and Cloud Reliability
The larger question concerns the specific cloud provider or platform you select. A cloud provider is a partner in the shared responsibility of maintaining an effective cloud solution. The provider should have no problem addressing questions on reliability and translating how their company’s features impact you.
In designing a cloud environment for your business, understanding what regions and availability zones to deploy your services, as well as different reliability patterns, is crucial, as it will have direct implications on costs, security, and ultimately, availability of services. Minimizing downtime is a primary concern of a cloud provider conducting proper maintenance on their end to sustain platform integrity. But I can’t stress enough that downtime is practically inevitable. Understanding the steps a cloud provider takes for regular maintenance and their guidance on cloud architecture best practices is critical to minimize the impacts of downtime.
I hope you enjoyed this series of posts on cloud computing principles! Reach out to me on Twitter, @SusanneTedrick for any additional questions or feedback!
Originally published at https://susannetedrick.com on September 28, 2021.
