In part 6 in my latest series showcasing the six pillars of the AWS Well-Architected Framework, we continue to take a look at the Performance Efficiency pillar. The Performance Efficiency pillar goal is to use a data-driven approach to operating as efficiently as possible covering resource selection, reviewing choices, monitoring resources and making trade-offs.
If you’d like to learn more about the other pillars of the Well-Architected Framework, check out the other blogs in this series via the links below. Otherwise, let’s get stuck in!
What we will be covering today
- What is Performance Efficiency?
- Common mistakes in performance efficiency
- Implementing good performance efficiency practices
Why we are learning this
- To help others navigate and understand how to improve performance efficiency across an organisation
- Using the AWS Well-Architected: Performance Efficiency Pillar for guidance to operate efficiently and effectively
How this will help me
You will:
- Understand good practices for operations covering people, processes and technology
- Be able to help champion performance efficiency across your organisation
- Understand the intersection between operations and performance of new services
What is Performance Efficiency?
Performance efficiency looks at how to run efficient workloads in the cloud. This means democratising availability of new services, building global architecture faster, leaning into serverless architecture, encourage experimentation and consideration of mechanical sympathy. The goal is to reduce operational overhead and free up resources to build and experiment with technologies that increase the velocity of building services that matter most to your business and clients.
Common Pitfalls of Performance Efficiency
Poor Selection
Many organisations struggle with service selection as it can be overwhelming with ways of achieving the same outcome. Where an existing pattern exists, it often results in choosing a “like-for-like” pattern. In migration this would be known as a lift-and-shift model. Modernising a legacy app is hard and the initial effort is often seen as not worth the trouble. This short-term and limited view often doesn’t take into consideration the cost/time savings that can be achieved in the medium/long term of using serverless architectures for example.
Missing or Long Review Cycles
Organisations rarely implement regular reviews of architecture to take advantage of new and emerging service capabilities. This often results in “tech debt” as solutions post deployment are left in a “legacy” way of working and managing. This adds to increased management overhead and leaves potentially more efficient ways of working in a suboptimal state.
Not Using Monitoring Data
Through leveraging data such as an effective monitoring strategy, organisations can make more informed choices about how to increase the performance and efficiency of architecture. There could be value in reducing operational overhead for example and reducing costs of provisioned resources by leveraging better auto-scaling capabilities of serverless architecture as a pointed example.
Limited Consideration of Trade-Offs
Understanding where trade-offs make sense and making pragmatic decisions around the use of new or emerging technologies drives additional efficiency. However this is rarely implemented or decisions made in isolation. This results in an autocratic implementation of technologies that stifle innovation and experimentation.
Implementing Good Practices
The term “best practice” is often used unassumingly. It implies a single way is “best” and this is the way things should be done. There isn’t such a thing as one-size fits all when it comes to each organisation, so flexibility in practices needs to be considered that are “good” for you. The below represents some good practices to increase performance efficiency.
Better Selection
Take a democratic approach to selection of services for your architecture. Thinking short-term will result in short-term gains. Take a longer term view to consider services such as serverless architecture is better for your architecture.
Regular Review
This is less about challenging convictions and ego’s as it is about consideration of what can be done better. Time is your advocate but leaving it too long between reviews can be detrimental to innovation. Regular reviews of architecture incites innovation and more willingness to experiment. This often leads to big gains in efficiency in the long-term and more often than not – a reduction in functional and physical costs.
Use Monitoring Data
An effective monitoring capability allows you to make data-driven decisions. This democratises the discussion about what services should be used in architecture by allowing the data to speak for itself and make factual decisions easier. Quantifying the gains can be useful here in political circumstances as you can explain the gains in terms of percentages as facts which increases the confidence of recommendations.
Well Considered Trade-Offs
Consider the trade-offs when using different technology to achieve a desired outcome. This sounds very obvious, however challenging the trade-off of not doing something can also be very beneficial. There may be several ways of achieving the same thing however more often than not, one of those ways will lead to better efficiency. This is where thinking big pays off.
