Digital Transformation Advisors

#1 – Digital Highway 2.0 or revisiting our Blueprint for Reliable Digital Solutions

At Swiss Digital Network, as one hub with multiple expertise, we have the joint purpose of enabling organizations and teams to deliver reliable digital solutions at speed and at scale.

This dedication to reliability stems from the background of the Swiss Digital Network founders, who spent more than 25 years helping teams to ensure the service levels, like availability, performance, and recovery, for business-critical applications. In early 2020, we designed a generic blueprint for reliable and continuous software delivery & operations, which we called the Digital Highway (see our blogpost May 2020).

In this article, we explain how DevOps and recently machine learning and AI have influenced the way digital solutions are built and operated today. Furthermore, the evolution of observability and the disruption in testing towards continuous verification are key drivers to rethink the Digital Highway.

The Roots of the Digital Highway

In the late 1990s ITIL (Information Technology Infrastructure Library) emerged as the reference best practices framework to help teams and organizations manage their IT services. Within ITIL, Service Level Management (SLM) was the practice dedicated to providing a framework to define and manage service levels with regards to e.g. availability or performance (typically as Service Level Agreements (SLA)). From the technical point of view, the SLM practice was implemented with systematic load testing and infrastructure monitoring at that time.

The adoption of Web legacy technologies such as JEE and .NET in 2000s drove the evolution to Performance Engineering (PE) and Application Performance Management (APM), as load testing and infrastructure monitoring in pre-production and production were not efficient enough anymore to detect and fix in time performance defects related to software design and runtime issues. In this context, we started integrating Performance Engineering and Application Performance Management into the existing process and tooling landscape to enable large-scale adoption and cope with the increasing adoption of agile development. Continuous Performance Engineering, Continuous Quality Assurance, and Unified Monitoring Architecture including self-learning Performance Management (the ancestor of AIOps!) were reference blueprints we developed to guide large customers in this journey.

The increasing adoption of DevOps, cloud, and the new opportunities offered by big data platforms, machine learning and AI, have pushed us to re-think our strategy to ensure the reliability of business-critical software and embed all reliability engineering activities into the new central infrastructure of software delivery and the first building block of the Digital Highway blueprint: the continuous delivery pipelines. Inspired by Google SRE concepts such as error budget, we merged them with our Continuous Performance Engineering and Unified Monitoring Architecture to build the  two other pillars of our first blueprint for the Digital Highway:

The three building blocks of the Digital Highway (Continuous Delivery, OOBASA, and Effective SRE) have been implemented in many customer projects and even applied to machine learning systems by our team at Machine Learning Architects Basel.

Almost four years later, after many real-live projects and implementations of the Digital Highway blueprint concepts, after leading joint conferences, and workshops with technology partners, and collaborating on research projects with academic partners, we concluded that we need to revisit and enhance the Digital Highway blueprint.

Our forthcoming blog post “Why Digital Highway 2.0? The need for Effective Engineering for MLOps & AIOps, Observability and Continuous Verification”, is scheduled for March 5th.

Leave a Comment

Your email address will not be published. Required fields are marked *