Let’s think of analytics, models, or applications as a ham and cheese sandwich. The fillings are biggest reason the sandwich is good. Let’s call this the purpose or value of the sandwich – after all, a sandwich isn’t just two slices of bread. And no desirable sandwich would have five or six slices of bread on either side before you get to the center.
Yet in IT deployments, this is exactly what we encounter time and time again. In fact, the concept of a sandwich in which the value is surrounded – and often discounted – by layers of redundant, repetitive technologies is an apt analogy for many IT deployments.
The past decade has seen a proliferation of fantastic analytical, application, and user value-based technologies. The opportunity to build more interesting “sandwiches” and the opportunity to mix and match has gone through exciting and profound changes. As the complexity of the enterprise has grown in customer base, markets, and geographies, internal demands for innovative solutions also have expanded dramatically, for the following reasons:
- Businesses are hungry to leverage the years of transactional data locked up in their ERP deployments.
- Conventional “storage and processing” technology followed the decades-long productivity curve, enabling efficient handling of enormous volumes of data.
- Sensors, far-reaching networks, and digital endpoints such as smartphones conspire to enable end-to-end analysis of personal shopper behaviors.
- “Digital business” is seen as an opportunity to outsmart competition, especially for non-conventional businesses.
- Effective, agile deployment of analytical solutions will become increasingly important to both revenue (competitiveness, market agility) and margin (cost of deployment, flexibility, and management).
We find ourselves highly challenged to create value across this environment of ever-increasing complexity because we still deploy solutions no differently than 30 years ago. To go back to the “sandwich” analogy, it may have many more and exciting ingredients, but for any of this to work, multiple layers of generic, repetitive “bread” – in the form of ETL, MDM, BPM, infrastructure, and data integration projects – have to be put in motion. So the constant need for these technical “slices of bread” dilutes the flavor or even the purpose of the whole sandwich. In essence, analytical, model, or application results – the value – are directly reduced by the number of slices of bread, i.e., the layers of integration and data access prerequisites.
Our industry is so enamored with the potential value of BI that we are ignoring the part that fundamentally undermines its value. Whether it was Oracle 1.0 nearly 30 years ago or Hadoop Cassandra now, the way we design, build, and deploy has not changed. Building analytics has become a little cheaper with bespoke development teams offshore, but the quality and satisfaction of the buyer continues to be low. And with a trillion-dollar backlog in application projects, something is fundamentally wrong with the current business value “sandwich” machine.
How can an explosion of heterogeneous sources of value and the resulting demands for fast, effective, and agile business solutions be delivered without the need for yet another wave of redundant, repetitive deployment and integration projects? In other words, how do we finally deal with the hard stuff? How can we reimagine the increasingly repetitive and redundant development and deployment models of the past so we can extract the value from a transforming and diverse topography of value?
The answer is heterogeneous analytics – the ability to mix and match diverse types of analytics in a continuous, linked, and highly productive fashion without the traditional overhead that creates siloed solution components with high degrees of “friction” between them. A further benefit of this highly productive paradigm is the ability to deploy ideas into solutions rapidly, to adapt to changes in the environment, and to maintain flexibility of one’s various assets.
Heterogeneous analytics helps businesses avoid “dead-end” investments that do not result in significant ROI. Focusing on the desired business outcome first and working back to technology investments drives a high degree of alignment and faster ROI. So the fewer slices of bread, the better the sandwich. And so it is with analytics and BI implementations. We are facing a massive change in the construct of the data market. To extract true value, the implementation and deployment model must be improved: reduce or maintain the present amount of bread while adding more meat. As a result, how we extract value and transparency from that market forces us to think of something new. Architectures need to align to the challenges they face rather than try to solve a challenge in a way that increasingly does not fit.
An emerging trend with terrific potential to accelerate heterogeneous analytics is microservices. This architectural approach addresses interoperability in the DNA of the design. Microservices are deployed as individually optimized nodes connected by fabrics of orchestration. Developers are able to encapsulate business logic into an environment that automatically inherits traditional data center disciplines like resiliency, availability, and security. The result is a simple “deploy and manage” solution leveraging an incredibly powerful orchestration fabric.
A well-designed microservices architecture allows businesses to focus precious development resources on solving the business elements of problems versus laborious conversion of logical design to opaque physical deployment. The more complicated elements (integration, performance, and resiliency) are simply handled within the fabric, or the combination of all these elements in a single harmonious construct. The results are spectacular: sandwiches with only two slices of bread – quick, easy, and tasty.
Simon Moss is the Chief Executive Officer and Board Member for Pneuron Corporation. In this role, Moss is responsible for overseeing the management, strategy and operations of the business on a global basis. Under Moss’ leadership, the company was founded, capitalized and the strategic platform developed, transforming Pneuron into a global technology provider. Simon brings over 20 years of successful strategic leadership at CEO, Partner and Board of Directors executive levels in the financial services industry, and has a proven track record as a successful entrepreneur.
Moss was most recently the CEO of Avistar and served four years with Mantas, moving the company into positive revenue streams and ultimate acquisition by Oracle Corporation. Previously, he served as Partner at Price Waterhouse Coopers, and was co-Founder of Risk Management Services Practice at IBM, all rounding out his extensive experience in financial services, trading and consulting practices. Simon excels in corporate revitalization and developing new go to market strategies for small emerging businesses as well as Fortune 1000 organizations. Moss is also on the Board of Directors for Qumas.
Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise.