Select Page

Post 1 – Outcome-Oriented Deliverables

It is a little bit surprising to me that so many people, even inside of IT, fail to understand what Enterprise Architecture truly is.  Most of the people I interact with think EA means Enterprise Technical Architecture.  They focus solely on the technical side of IT/EA and miss the big picture for what EA truly is.

Many other individuals get so caught up in the process of doing EA (ie creating artifacts) that they fail to realize what is it they should be doing and should be accomplishing.  Every decision that gets made needs to be able to answer the question “how does this evolve the business and enable the senior executives to respond to disruptive business threats and opportunities.”

The architecture and the modeling language selected by the EA practitioners need to be valuable to senior management and need to be able to direct change and empower business and IT leaders to make better decisions.  The technology chosen for EA use needs to enable enterprise architects, solution designers, program/project managers, or business managers, to analyze the impact of different decisions.  In doing so EA practitioners can link EA and business strategy.  These deliverables created to drive the business forward.  Gartner identifies five types: measurable, actionable, diagnostic, enabling, and operational.

  • Measurable Deliverables: Measure the direct impact of EA on the business.  They can take two forms: Quantitative metrics and Qualitative metrics.
  • Actionable Deliverables:  Initiate change through projects, or guide projects with specific principles and standards.
  • Diagnostic Deliverables:  Combine different views of a problem or opportunity to address a specific need.  These may include technology evaluations, future-state models, and assessments of current-state business.
  • Enabling Deliverables:  Uses collective information to provide input to diagnostic deliverables that represent the business, people, processes, information, and technology.  This can be used to help advise IT and business management when making a decision.
  • Operational Deliverables:  These are artifacts used to define and run the EA program.  They have little business value as they’re largely focused on what the EA practitioners do and on positioning EA goals and governance structures.  Examples might be the EA team charter and EA steering committee.

References:
Burke, B., Burton, B. (2012). EA Practitioners Must Focus on Outcome-Oriented Deliverables. Gartner.

 

 

Post 2 – Bimodal Infrastructures

The infrastructure for many modern enterprises, including where I work, is quickly evolving from hardware-centric silos to software-driven ecosystems.  This new wave of application architecture is quickly changing how our infrastructures are developed and utilized.

With this new application architecture having such a big impact on the infrastructure architecture the infrastructure architecture is being reshaped in terms of its compute, virtualization environments, and software ecosystems in order to ensure a simple, streamlined process of efficient delivery that can quickly be learned by staff supporting the digital business.

If today’s digital business doesn’t learn to embrace the new application architecture and shift the resources away from hardware-centric IT towards developing environments, deployment models, Mode 2 workloads, and the overall agile, non-sequential, experimental digital business than they will quickly face irrelevance.

Bimodal infrastructures will need to become the de facto standard as enterprises shift from the traditional mode 1 approach, which historically utilizes a more on-premises compute infrastructure, to a mode 2 approach.  This approach will give the enterprise reliability and agility from both modes as the business shifts to emerging technologies and externally sourced compute.  Businesses should take the physical infrastructure from a set of disaggregated components and create pools of resources form which to configure applications.  These containers can drive greater compute efficiency and agility and provide a way for developers to modernize their applications without having to consider computing resources.

With many large and complex IT organizations the infrastructure has historically been inflexible and less than agile.  This has often resulted in inefficient workflows, poor cross-team collaboration, and process that inhibit projects from getting up and running quickly.  Nowadays it’s important for leaders to shift their priorities to build closer ties with the business rather than continuing to invest in traditional on-premises hardware-centric data center technologies.  Building a bi-model platform and utilizing public cloud and SaaS offerings will help the business to deliver projects quicker and with more confidence in their performance.

This shift in priorities breaks down the component silos that has become the norm for many businesses as compute infrastructures have grown and evolved over the years.  Removing the physical elements altogether collapses these infrastructure silos and allows for ‘pools of resources’ to be developed.

To start, leaders need to invest in infrastructure technologies that reduce management cost and complexity.  They need to collaborate with business and application leaders to ensure their visions align.  During this time it is important to ensure your team consists of individuals who have knowledge of private and public cloud technologies.  These individuals should be used to begin deploying applications in the cloud.  As more applications move to the cloud silos can be shut down, thus freeing up resources within the business.

References:
Cisek, M. et al. (2017). 2017 Strategic Roadmap for Compute Infrastructure. Gartner.

 

Post 3 – Enterprise Data Centers

With the presence of the public cloud increasing year after year on-premises data centers are being re-evaluated in almost every industry.  However, organizations have many services they offer that utilize on-premises data centers and can’t immediately utilize cloud options, and perhaps shouldn’t depending on the organizations need.  With this in mind, how do organizations leverage the data centers they currently have when over time fewer services they offer will continue to utilize it?

The digital business if forcing IT to rethink its service delivery model.  IT needs to begin thinking about how a service should be delivered to address the needs of the enterprise.   This thought process needs to permeate to the infrastructure and operations teams in their decisions regarding the data centers and services they provide.  The I&O team will need to leverage technologies that enable a software-defined data center.  The decisions they make need to be based on service delivery and business expectations, not on hardware or implementation details.

The public cloud will not be a panacea to solve all IT problems, but the I&O team should investigate which type of cloud service (IaaS, PaaS, SaaS, etc) and which provider (Amazon, Google, Salesforce, etc) will help the enterprise efficiently and effectively exploit the services offered within the business.  The I&O team needs to be focused on the services that need to be delivered and again, not on the hardware, software, architecture, network, etc.    Which services add value to the enterprise?

An Enterprise-Defined Data Center (EDDC) needs to recognize its role as an intermedia in a multi-provider ecosystem.  The EDDC will be brokers of the services rather than solely being providers.  IT is no longer about keeping hardware up and running, which is something I know many people in the field have difficulty grasping, but it’s about recognizing the direction the organization is heading and working with the business to utilize technology to help the business achieve its desired goals.

There are a few tools out there that can help EDDC utilize its data centers efficiently.  Some of these are:

  • Enterprise Workload Monitoring (EWM) – track applications and services and support workload dependency mapping across public and private infrastructures
  • Global data center management (GDCM) – monitor and manage computing assets and associated workloads and costs across owned data centers, colocation, and hosting providers, and cloud service providers.
  • IT operations management (ITOM) and IT asset management (ITAM) – provide IT operations control of on-premises equipment and applications.

With EDDC, these tools can help the IT teams manage their services attributes, performance, KPIs, and cost, without concern about where the services are physically hosted.

References:
Cappuccio, D. J., Bittman, T. J. (2015). How to Grow the Enterprise-Defined Data Center. Gartner.