Last month Scott Moore hosted the Atlanta stop of his Performance Tour at Shadow-Soft HQ. In this blog post, Shadow-Soft’s Bill Hayden shares a recap of the key ideas discussed during the event.
Scott Moore and I have had the privilege to watch firsthand how the world of application performance has changed over the past 20+ years. Each tectonic shift in application architecture and technology has brought new challenges, and now these shifts are increasing in frequency leaving us sprinting to catch up.
From client server to web, from web to SOA, from SOA to microservices and now on to containers, cloud native and serverless, each technological evolution has impacted application performance in unique ways. Conducting performance testing on Prod-like environments has given way to intelligent Application Performance Monitoring in Production with a laser focus on minimizing mean time to identification and resolution.
At the same time, end users, particularly for mobile applications, have little tolerance for excessive response times and often the vast majority of those delays are introduced by the network. This leaves App Dev teams little choice but to design with millisecond budgets in mind.
The purpose of DevOps is to deliver code at velocity which is very much in keeping with the ethos of performance. However, designing performant code is not always at the forefront of mind. This is in part due to the ability of Cloud providers such as AWS and Azure to scale horizontally as well as vertically which has allowed systems to meet customer demand and expectations even without performant code.
Rarely though is the question asked why scaling was necessary in the first place. However, with the advent of serverless and the direction that Cloud billing models are taking, the cost of computing may soon be directly linked to application performance, creating a new dynamic that will make this discipline more important than ever before.
Touching all of these considerations is the concept of Site Reliability Engineering, particularly for containerized, cloud deployed systems. Platform reliability and system resiliency will now be in the forefront along with performance considerations, especially in the era of containers. This will in turn require an SRE mindset and an approach to Kubernetes that avoids technical debt while accelerating cloud deployments. As a result, an effective Kubernetes Container Platform, like Red Hat OpenShift, must be the cornerstone of any application modernization initiative.
Performance Tour Ep 13 – Atlanta
In episode 13, Scott Moore concludes the Season One Performance Tour in Atlanta, GA at a local meetup event hosted by Shadow-Soft. He discusses current performance trends and samples Spiced Right Rib House BBQ.
Kubernetes and OpenShift Consulting Focused on Enablement and Integration
OpenShift is your fast pass to the cloud. We can help you configure, build, and manage an enterprise-ready Kubernetes platform with Red Hat OpenShift.
Speak with our Technical Sales team to learn how we can help your organization with implementation and enablement services.