As organizations plan, migrate, transform, and operate their workloads on AWS, it’s vital that they follow a consistent approach to evaluating both the on-premises architecture and the upcoming design for cloud-based architecture.
To guide enterprises in their evaluation process, AWS has developed the Well-Architected Framework that describes key concepts, design principles, and architectural best practices for designing and running workloads in the cloud across five areas, or pillars: operational excellence, security, reliability, performance efficiency, and cost optimization.
Well-Architected Framework design principles include:
- Using data to inform architectural choices and improvements over time
- Fully conceptualizing capacity requirements
- Tracking changes to automated processes, including auditing impacts to the system, and reverting to the previous environment states seamlessly.
- Allowing architectures to be nimble and evolve over time, allowing organizations to take advantage of innovations as a standard practice.
Well-Architected Reviews are conducted by AWS customers and AWS Partner Network (APN) Partners to evaluate architectures to understand how well applications align with the multiple Well-Architected Framework design principles and best practices. The ultimate goal of each of these reviews is to identify gaps, quantify risk, and develop recommendations for improving the team, processes, and architecture with each of the five pillars.
Well-Architected Reviews identify gaps and risks that depend on the business priorities, architecture, and organizational culture and processes for the workload being evaluated.
Below are some high level concerns we hear in the industry:
- Development teams design for happy paths. Failure paths are thought of too late in the lifecycle. We are concerned about the lack of transparency to our applications and the risk this brings to solving operational issues.
- Operations and Resiliency Monitoring is still waterfall and not designed into our applications. Timelines are getting missed because monitoring is being put in at last minute.
- When unexpected things happen (security threat or reliability issue), we don’t have the right signals and plan to remediate them quickly
- We don’t have the right set of tools in place to understand the health, validate architecture decisions and ensure we are delivering on all business expectations.
- Not only do we struggle to quickly deploy new services and the underlying IT infrastructure, we also don’t know if we are disrupting our end users until they complain.
- We lack the baseline for current workloads to analyze the costs as we transform to all-cloud or hybrid-cloud topologies.
To bring higher-quality information to Well-Architected Reviews and to establish a strategic advanced observability solution to support the Well-Architected Framework 5-pillars, Dynatrace offers a fully automated, software intelligence platform powered by Artificial Intelligence.
With out-of-the-box, zero-configuration, Dynatrace immediately lights up your environment providing the right insights to understand your application heath across the five pillars and to quantify risks identified by the Well-Architected Review.
Below are a few examples within the five Well-Architected pillars for how Dynatrace empowers you:
Dynatrace and AWS
Dynatrace provides advanced observability across on-premises systems and cloud providers in a single platform, providing application performance monitoring, infrastructure monitoring, Artificial Intelligence-driven operations (AIOps), code-level execution, digital experience monitoring (DEM), and digital business analytics.
Dynatrace is an advanced AWS technology partner, and together with AWS, we help enterprises stay on top of their complex, dynamic, and scaling AWS cloud environments through our AWS integrations and monitoring support.
Here are a summary of the growing list of Dynatrace integrations for AWS:
- Full-stack monitoring of EC2, Lambda, Fargate, and Elastic Kubernetes Service (EKS) for a deep understanding of the vertical and horizontal dependencies across your on-premise and AWS cloud environments using Dynatrace’s OneAgent and Smartscape
- Easy deployment of Dynatrace OneAgent with AWS Systems Manager Distributor, AWS Elastic Beanstalk, and AWS CloudFormation.
- Automatic collection of the entire set of services that publish metrics to Amazon CloudWatch. (these metrics are also automatically analyzed by Dynatrace’s AI engine, Davis®).
- Seamless monitoring of AWS Services running in AWS Cloud and AWS Outposts.
- Dynatrace as a managed AWS workload, and as an option, have the network traffic to Dynatrace run over PrivateLink so that traffic never leaves AWS.
- Automatic transfer of Dynatrace AI-detected problems (including affected instances and related events) into AWS services with AWS AppFlow data transfer service.
We believe that Dynatrace and AWS together provide the answers required to understand workload health during Well-Architected Reviews, validate architecture decisions, and helps serve as the observability platform that complements and extends AWS monitoring.
How to get started
Solving operational issues has be designed into your application on day 1. Without that timelines get missed when monitoring has to be put in last minute. Well-Architected Reviews reviews force the conversation early, combining that with a well-defined execution mechanism will prevent concerns around transparency of the applications.
As you embark on the AWS Well-Architected Reviews or are now looking for how to address risks and shortcomings identified from your review, sign up for a free 15-day trial here. Or, if you’re an existing Dynatrace or AWS customer, talk to your Account Executive or Customer Success Manager to learn how Dynatrace or one of our highly skilled and experienced solution partners can help you master the AWS five pillars.
If this piqued your interest, stay tuned for the second part of this Well-Architected series where I’ll explain how the data gathered with Dynatrace relates to the five pillars of the Well-Architected Framework.