TOPIC 4 – Technology Architecture

Post 1 – Technology Architecture in Global Organizations

Most multinational organizations are dealt with the complex challenge of dealing with local laws and local culture/business practices when setting up shop. Couple this with a technologically challenged local workforce and you have architecture problem brewing.

Since most of my blogs are based on personal experience, I’ll list out another example of the importance of Technology architecture alignment and its benefits. We will then look at how these benefits are sometimes ignored due to a myriad of reasons.

Speaking purely from lessons learned & educational inputs received from diversified experiences; most Technology architectures are results of from two critical elements:

• Vision
• Ability to execute

Let’s look at that in more detail.

The enterprise vision is clearly articulated and known by the time we get to Technical Architecture decisions. From the TOGAF diagram above; we can see that Technical architecture decisions are based on exercises completed in Business, Data & Applications domain. At this point, all stakeholders must know :

• Core business processes that run the organization
• Data Models and matrices
• Application architecture that support business and data processing needs have been identified

The Technology architecture decisions made must be based on inputs received from all the above-mentioned  architecture domains. TOGAF provides the below steps to define Technology architecture. The next phase; Opportunities and Solutions (Phase E) is ideally where all the (Application & Technology) decisions are made.

Before we get to the second element (Ability to execute) lets look at what happens in reality.

Many organizations don’t start with a well defined architecture definition process or even an acute awareness of such a need. They focus on being profitable, increasing their customer base and achieving organic growth. As the organization moves from fighting for survival to a business with steady incomes, large customer base and rising profits; the focus shifts to optimizing the applications & technologies that were cobbled together in the past when resources were tight. If not addressed effectively; this leads to the first of many potential problems that the organization will face over its existence. You will find many organizations with “legacy” systems that continue to run in parallel to core business applications. In most cases; this is a result of stakeholders with a reluctance to let go of old/redundant systems or political maneuvering. The organization now spends resources maintaining redundant systems and faces lower ROI.

Now; if you were to take the above breed of systems and apply it to a scenario where the organization plans to setup shop in a new country; you’ve got a training, execution and optimization problem. For example, the sales system purchased initially may not have been designed to accommodate additional currencies, exchange rate conversion, local tax types & calculations, etc…The stakeholders are now left with an option to implement a new sales system and/or customize the existing application to accommodate their needs; and this can get expensive very quickly.

Which brings us to the next element; the capacity/ability to execute.

If we were to take the lessons learned from the sales system example i just described; there are two high level potential outcomes:

• The organization learns from its mistakes and decides to streamline its efforts going forward by deploying architecture (business, data, application & technology) decisions that are strategically and globally relevant. i.e. globally capable sales, production, logistics capabilities
• The organization fears a repeat of the colossal disaster it faced the first few times and limits its risks by allowing local (country based) stakeholders to make their own decisions. This is one example of where extensive discussions on integration architecture are triggered every time a country decides to revamp any one solution

There isn’t much of a conclusion to reach here. These are observations that anyone can make when trying to understand why an organization’s technical architecture is extremely competent or otherwise. For most organizations that are in this situation; the approach must be to make conscious efforts to align to a global, unified architecture & capability model.

References:
– TOGAF Technology Architecture definition

– Robertson, B. (2007). Defining a Technical Pattern Starter Kit, Gartner. August.

POST 2 – Technology Architecture in the Cloud

in my previous post, we talked about technology architecture and the limitations organizations face with deployment while being cognizant of the precautions one needs to take to ensure success. The proliferation of cloud based technologies have made organizations sit up and notice.  The challenge however; is the deployment of cloud as a technology platform and using its capabilities to deploy applications.

Application and Technology architecture used to be spoken about separately in the past. However; platforms like Amazon Web Services, Salesforce.com, Microsoft, SAP, etc…. now create a grey area on applications and technology. While this is a good problem to have for organizations looking to cut costs and optimize resources; the approach to cloud deployments must be measured and executed.

Gartner suggests aligning your cloud infrastructure strategy in 4 steps:

• Define Scope by evaluating the long term benefits of private and public clouds, the benefits and challenges of adopting emerging technologies
• Align to/or Develop IT Strategies to determine if this approach makes sense to the organization’s long term goals.e.g. does Big Data & Live Data Analytics a necessity for an organization engaged in short term, non-profit exercises?
• Define strategic plan for Cloud infrastructure – where you determine a lift & shift approach or a hybrid approach is ideal
• Create Operational plan for infrastructure – The final execution plan to move the organization from current baseline of on-premise technologies to a cloud based industry leader

References:
– TOGAF Technology Architecture definition

– Warrliow, M. & Gill, B. (2017), Align Your Cloud Infrastructure to Your Business Strategy in Four Steps … or Fail, Article G00325899

– Cearly, DW, (2017), Devise an Effective Cloud Computing Strategy by Answering Five Key Questions, Article G00270415

POST 3 – What’s next for cloud architecture

I happened to read an interesting article on what’s next after cloud technologies have matured. The article calls out the limitations with cloud technologies, namely:

• Latency – While latency has improved with time using web technologies and HTML5 based applications, latency continues to be an issue. The best of domains and service providers continue to operate in the realm of 3-5 milliseconds and are nowhere close to solving the issue of latency.
• Bandwidth – We continue to see improvements in fibber optic technologies and 5G pilot deployments. However; the progress of bandwidth development isnt keeping up with the pace at which data is being generated by IoT devices on a daily basis. This can potentially choke a data pipeline in case of a global or local event. Here again, cloud based technologies are limited in their ability to address critical business needs when they’re constrained by bandwidth issues.
• Autonomy – The need for complete and self managed automation on cloud hasnt matured yet. While there are improvements in AI; these are nowhere close to achieving self correction or remediation needs.
• Privacy – We’ve seen multiple AWS S3 buckets leak data just in the recent 2 years alone. While most of these could have been re mediated by well managed configuration; the lack of knowledge and potential impact simply shows we have a long way to go.

The concept is fairly simple. Over time, newer technologies will eat up incumbents.

The recommendation here is to keep an eye out for improvements in localized data processing and compute capabilities i.e. faster mobile devices and laptops.

References:

Bittaman, J.T (2017), Maverick* Research: The Edge Will Eat the Cloud, Gartner research article G00338633