Notes from the Field: Diversity of Conversations, Hyperconvergence, and Adaptive Enterprise

One thing can be said about the world of consulting: the conversations you have with customers never cease to be diverse. My current engagement with a large multinational enterprise that contains multiple business areas with multiple business units is the epitome of that descriptor. We go from high-level, esoteric conversations about economics, value exchange, and business models, to deep in the weeds technically, such as hyperconvergence, and then back to the Middle Earth that is the analysis of technologies that can help achieve their goals.

One of those conversations recently was around the term “hyperconvergence” and how it figures into the design principles and rules that I laid out for the company back at the beginning of the engagement. Just as a refresher for you, those principles and design rules are:


  1. Modularity
  2. Integration
  3. Standardization
  4. Simplicity

Design Rules

  1. Virtualization
  2. Model-driven architecture
  3. Service-oriented architecture

A slightly deeper dive on the design principles:

  • Modularity: Minimize the dependencies between changes. Modularization is the process by which monolithic structures are broken down into reusable components. A module provides a stable interface to a set of functionality whose implementation is hidden so that it may be changed (or substituted) without impacting users of the module. The fundamental unit of modularity in the adaptive enterprise is the service.
  • Integration: Enable the composition of separate modules into useful systems. The service-oriented architecture (SOA) is the primary integration method in the adaptive/extended enterprise and is used to link business processes and heterogeneous IT resources dynamically.
  • Standardization: Facilitate integration, maximizing reuse, and extraction of value. The method used to handle heterogeneity in the data center, standardization includes the use of emerging industry-standard operational reference models for business processes and the application of industry-standard interfaces and platforms. This adaptive/extended enterprise approach standardizes both IT service delivery systems and IT service management systems.
  • Simplification: Minimize what needs to change and associated costs. The primary remedy for dealing with IT complexity, simplification includes the consolidation of applications and infrastructures and the automation and orchestration of processes based on service models.

The view of the adaptive (or as they like to call it now, extended) enterprise emphasizes the organization of IT into a set of managed, layered services that deliver reliable business services in alignment with the overall enterprise’s business objectives and strategies.

Extended/Adaptive overview
Extended/adaptive overview

Design rules are design decisions that are consistent with the principles. They are fundamental, well-defined prescriptions that allow for nearly independent innovation. They define the flex points of the architecture, the keys to delivering on the adaptive promise. If the design rules are understood and respected, architects can define solutions that will evolve gracefully and enable progress toward the adaptive/extended enterprise one solution at a time—and, in the case of this most recent conversation topic, “hyperconvergence.”

Because the concept of the adaptive/extended enterprise addresses a range of concerns, many different views of the architecture are useful. Two of the primary views are the enterprise view and the IT services view. The enterprise view provides a holistic model of the overall enterprise and its extended value chain; the view includes:

  • Overall business objectives and strategies
  • Business processes for delivering services and goods
  • People, technology, and other assets for executing the business processes

The enterprise view is critical for understanding IT’s role in the enterprise and how business and IT synchronization can be accomplished. A formal business process model that includes KPIs is a key element in the synchronization. Note that the enterprise view is a general model of any organization. As such, it can be applied to organizations of lesser scope than the whole enterprise: in particular, the various business units and the IT organization itself. The IT services view provides a model of the core IT capabilities that are used to provide information services to the enterprise and its stakeholders. IT services are organized into three major layers:

  • Business services, which automate the business processes that provide business services to the enterprise’s customers
  • Application services, which are the building blocks for the business services
  • Infrastructure services, which are the core building blocks for the application services

This view also shows how the IT service delivery systems are managed by the related service delivery management systems at each of the three layers. The design rules—SOA, virtualization, and model-driven automation—are key for the IT services view.

Now that the review is complete, let’s get back to the “hyperconvergence” topic of conversation. I asked for customer’s definition of the term “hyperconvergence,” and in typical fashion, I received four somewhat different responses. So, I used the definition that fit the situation from Scott D. Lowe, “hyperconvergence is a way to enable cloudlike economics and scale without compromising the performance, reliability, and availability you expect in your own data center.” Pretty generic without diving into the different solutions that are out there, but good enough to give us something to work from as we align this with some of the design principles and rules.

Let’s start from the top: modularity. Since the topic of conversation with this customer was, in general, “hyperconvergence,” the solutions that fit within that definition are primarily in the infrastructure services sections in the graphics above and following:


Modularization breaks down monolithic structures and creates reusable components and layered, abstracted functionality to improve sharing and reuse. The very nature of hyperconverged solutions delivers infrastructure services simply by consolidating all required functionality into a single infrastructure stack (we could say “modules” here as well) running on an efficient elastic pool of resources. The graphic above also shows virtualization, which is closely related to modularization. As we all know, it organizes components into uniform, logical services, making the components interchangeable (substitutable) and shareable. And when a resource is needed, virtualization can cause the resource to be created or allocated using available lower-level resources with similar characteristics.

Integration: With regard to the current hyperconverged solutions on the market today, it’s safe to say that the solutions provide sufficient integration across both management and execution environments as it specifically relates to the infrastructure services and control layer. These solution providers have established a management integration platform whereby enterprises can implement a loosely coupled, service model–based, heterogeneous integration platform for management functions. In addition, enterprises can instrument hardware and software resources to ensure management has both the information needed to make decisions and the mechanisms to put them into effect.

With hyperconverged solutions, this customer can set explicit expectations regarding management characteristics, such as interoperability, availability, performance, security, scalability, mobility, and maintainability. One area of high interest with my customers is evolvability or agility: that is, the ability to adapt their IT’s capabilities over time to meet new requirements. In other words, IT must be able to integrate incremental capabilities and consolidate existing systems.

In general, by integrating within and across the execution and management environments, enterprises can leverage hyperconverged solutions to achieve a competitive advantage and improve agility.

Standardization is the method used to handle heterogeneity within IT. It allows systems to interoperate and also allows the creation of interchangeable components and processes. Standardizing interfaces, tools, service templates, and processes simplifies the environment and provides economies of scale in purchasing, usage, and management; reduces cost and time to implement and integrate; and reduces spending on professional skills. Hyperconverged solutions fit perfectly from the standardization perspective in the infrastructure services layer.

Simplification: Simplification is the primary remedy for dealing with IT complexity. Simplification is a key conversation topic with this particular customer and one that really strikes home when we talk about hyperconverged solutions. Approaches to simplification include the following:

  • Use of models to automate service delivery and management. Focus on processes that are repetitive and prone to human error, and work toward applying closed-loop control to allow self-managed, self-healing, and self-optimizing systems. This will save time and money, reduce human errors and risks, and increase quality.
  • Consolidation of applications and infrastructure by reducing the number of types, versions, and instances. Enterprises can then focus resources on a smaller set of components and allow economies of scale and management.

In today’s ever-changing business climate, successful companies are those that create a hyperconverged solution-based adaptive/extended enterprise that can effectively synchronize business and IT. By matching their evolving business needs to their IT environments, they can strike a balance across managing costs and risks, increasing value and quality, and enhancing business agility.

In these hyperconverged solution-based adaptive/extended enterprises, the IT organization is transformed into a fundamental enabler for the enterprise. Through this transformation, the enterprise shifts from a series of disjointed business units with isolated, manually connected applications to one that has distributed, integrated business processes; connects with trading partners; and is supported by a common, shared infrastructure. This transformation eliminates stranded capacity and introduces a new flexibility that enables the enterprise to respond quickly to change.