#3 – Digital Highway 2.0 or the five Engineering Pillars to enable Reliable Digital Solutions 

In this revisited version, we define the Digital Highway as a blueprint for high-velocity delivery of reliable digital solutions, based on five engineering capabilities: 

1. Effective SRE for software, data, and ML models 

2. Effective Continuous Delivery (CD) for software, data, and ML models 

3. Effective MLOps, AIOps, and DataOps 

4. Effective Observability for software, data, and ML models 

5. Effective Continuous Verification & Testing for software, data and ML models 

Furthermore, we strongly believe that designing, building, and maintaining the Digital Highway capabilities requires: 

• The adaptation of the operating model to shift and scope the responsibilities between existing and new engineering roles & engineering skills, to enable newly assigned staff to fulfil their new duties related to SRE, CD, MLOps, observability, and continuous verification. 

• An appropriate mindset fostering shared responsibility, people centricity and error-friendly culture. 

As in previous blogs, the concepts related to effective SRE, Continuous Delivery, and MLOps have been introduced and illustrated; we will give a brief definition of these 3 engineering capabilities and add more details to introduce the new scope and definition of Effective Observability engineering and Effective Continuous Verification Engineering. 

Effective SRE

• The application of a systematic, holistic, disciplined approach to the definition, evaluation, and cost-effective assurance of Service Level Objectives (SLO) Engineering in the context of the Digital Highway (through the CD pipelines and in Operations). SRE is usually required as a central capability in Digital, DevOps, or Cloud transformation roadmaps.  
• Effective SRE practices can be tailored to address Data and Model pipelines in addition to Software pipelines. 

Effective Continuous Delivery (CD)

• The application of a systematic, holistic, disciplined approach to cost-effective, reliable delivery and release of components through a pipeline of different environments and stages, balancing speed and quality. 
• It harnesses artificial intelligence and Machine Learning capabilities to drive seamless quality verification and orchestration of continuous delivery pipelines for Software, Data & ML Models 

Effective MLOps

• The application of a systematic, holistic, disciplined approach to a cost-effective delivery of reliable Data-, ML-, and AI-driven systems.  
• Effective MLOps applies advanced DevOps practices such as SRE, Continuous Delivery, and Observability to Data- and ML-Pipelines, in addition to the Software pipelines. 

Effective Observability Engineering

We define Effective Observability Engineering (EOE) as follows: 

• The application of a systematic, holistic, disciplined approach for the design, instrumentation, and cost-effective data collection, processing & analytics to observe and understand the behaviour of complex digital systems and increase their operations & development productivity. 
• Observability Big Data are required to efficiently collect, store, and analyse high volume of observability data that are relevant to different use cases such as monitoring, testing and QA, security, change management or DevOps DORA and Value Stream Management (VSM) analytics. 

The Effective Observability Engineer is responsible to: 

• Co-Build, Maintain & Operate O11Y-driven appl. Instrumentation code  
• Co-Design & Select or build the O11Y Big Data 
• Co-Build, Maintain & Operate O11Y-driven Analytics & ML-Toolkit * 

*The O11Y-driven Analytics & ML Toolkit is required to enable an automated interrogation of the O11Y big data for multiple use cases such as SLO Engineering, Troubleshooting or increased DevOps productivity 

Effective Continuous Verification

We define Effective Continuous Verification (short CV) Engineering as follows: 

• The application of a systematic, holistic, disciplined approach for the design, instrumentation, and cost-effective data collection, processing & analytics to observe quality gates and system resiliency through the delivery pipelines for software, Data, and ML models, automating their Verification and increasing the overall testing & SRE productivity. 
• CV is always ML-driven and interacts with Observability Big Data, which provides the required quality, SRE, and experimentation (Chaos Engineering) data. 

The Effective CV Engineering is responsible for: 

• Co-design, Model & Implement the Data Model and ML-driven Decision-making Algorithm to automate QG Verification 
• Co-design, Model & Implement the Data Model and ML-driven Decision-making Algorithm to automate the Verification of system behaviours under unexpected conditions (Chaos Engineering) 
• Co-design, implement and maintain an ML/AI-driven tooling chain to improve Testing & SRE Productivity 

Impact on the Digital Highway Operating Model and Culture

Like leading technology vendors and senior engineers from global digital players with highly advanced DevOps Maturity, we consider the above-introduced engineering capabilities of the Digital Highway as an advanced DevOps engineering practice whose adoption requires a holistic culture-first transformation in addition to technology and operating model. 

However, most teams and organizations struggle with the cultural dimension and take major risks with technology-only initiatives. Keeping the conventional hierarchy-driven culture & mindset is not efficient for handling the speed, challenges, and complexity of the highly distributed and changing digital systems and might be considered a major barrier to the success of the new engineering practices such as SRE, Observability or MLOps. Furthermore, top-down culture transformation approaches are unsuccessful because they lack authentic adoption by the “people.” 

At Digital Transformation Advisors, our Senior consultants, based on their proven track record in IT Transformation projects and driven by their engineering spirit, developed a Culture Maieutic approach inspired by the concepts 1) Community of inquiry (https://en.wikipedia.org/wiki/Community_of_inquiry)  and 2) the Socratic maieutic to help teams collectively shape their new collaboration value charter and culture. 

Practical Maturity Models will be used in addition to enabling the teams to collectively identify their current assets and gaps and define their target state in terms of the 5 Digital Highway engineering capabilities: Effective CV, Effective Observability Engineering, Effective MLOps, Effective CD, and Effective SRE.