AWS Well-Architected Cost Optimization Labs Updated
AWS’s breadth of resource options and purchase models allow you to design your infrastructure in a most cost-efficient way possible. You can start by identifying idle and underutilized resources with services such as AWS Cost Explorer Recommendations, matching your resource capacity with business demand using services such as Instance Scheduler, and taking advantage of discounted purchase models based on your workloads’ stability and fault-tolerance level. With the experience of reviewing tens of thousands of customers’ workloads, AWS Solutions Architect team designed the AWS Well-Architected Cost Optimization Framework and Lab. The cost lab provides a hands-on environment where you can get familiarized with some of the most commonly used cost management tools and best practices. In this blog post, Nathan Besh, our Principal Cost Solutions Architect, shares the latest of our Cost Optimization Lab.
We recently refreshed our Well-Architected Cost Optimization labs with new content and a new look. AWS regularly releases new features and updates to our services, so we periodically update our labs to make sure they are accurate, and also the most efficient way to demonstrate how to implement the best practices. These updates can be a simple screenshot when the console has been updated, or removal of an entire lab when it’s no longer relevant.
We have implemented Hugo, which is widely used across AWS for workshops and labs, so you can have a consistent learning experience. Hugo also includes a search capability, making it easier for you to quickly find the right lab and content. There are also some visual improvements, most notably in the use of the page space. There’s more space dedicated to the actual content and less space consumed by menus and other fields, so larger images, less word wrap, and less scrolling!
Our services and features continually change, so we regularly update our labs to ensure that they show the current console experience. We have refreshed the cost labs to include new screenshots where there have been changes across our services.
Streamlined lab experience
We have been building out our Well-Architected labs, one at a time, over the last two years. We aim to have relatively short labs, which should take about 20 minutes to complete. Now that we have over a dozen labs, we examined ways to simplify the experience, such as making newer labs build upon the previous ones to eliminate redundant, and often tedious work. A great example of this is the new account setup lab, where we made a single modification to the Amazon QuickSight IAM policy that removed 10 steps every time you create a dataset in QuickSight. Previously, these steps were repeated across three different labs.
Updating the approach to Cost Optimization
Historically, AWS used a payer and linked account structure, where a payer account was used to pay the bills for any linked accounts. With the introduction of AWS Organizations, this payer account—now known as the Master account—has a significant role in the security of your AWS accounts. Cost Optimization has also grown as a domain in the industry, with dedicated staff and full-time roles. This growth has driven the need for resources specific to Cost Optimization, such as dashboards and analytics systems.
With these changes, we now take a different approach in the labs by creating a specific member account for Cost Optimization. The Cost Optimization team uses this account to perform Cost Optimization activities, and only have limited access to specific services in the Master (payer) account. The Cost and Usage Reports (CUR) are now set up to support cross account access. We’ve also changed from using IAM users, groups, and roles to AWS Single Sign-On (SSO) with permission sets. This arrangement allows central management of access across multiple accounts.
We will be adding new labs to ensure that you have the practical hands-on guidance you need to build capability and ensure that you are well architected. Check out the updated Cost Optimization labs and be sure to come back often. We’ll be adding new labs for best practices, with a focus on high risk items, and extending the existing labs on Savings Plans. Also expect to see us add labs on non-technical best practices, such as developing goals and targets.