retrospective late binding and m&a
A datagenous client went through the following situation:
- A 16.8B acquisition across 75 locations in 19 countries with 17,000 worldwide personnel.
- A concurrent 2.75B spin-off including 11,000+ employees, 85B+ devices with 10,000+ parts.
- Preparing to be acquired for over 40B by a third party
Executing complex IT transformations to realize the business strategies behind these events required a data and IT service strategy that could address:
- Disparate, dynamic, environments (complexity)
- Multiple concurrent reorganizations (uncertainty)
- Competitive pressure to identify and deliver synergies (speed)
the datagenous process
- scope - identify target system schema and data types
- project - define initial control flow, construct initial graph/schema/model
- normalize - concurrent data resource acquisition and access normalization
- define system data flow - identification of minimal viable implementation (MVI)
- retrospect - iteration of schema, control flow, rules, and feedback loops - definition of release candidate qualification criteria, (RLB)
- validate control flow - identify release candidate, beta
- deploy - release RC to production
- optimize and scale - consolidate, federate, or distribute supporting services based on performance and scalability criteria and changing infrastructure conditions
Using the datagenous service our client was able to:
- create the data services to consolidate views of the product life cycle management systems in 6 weeks with a team of 2 experts
- execute the spin-out in 3 months
The datagenous service made these outcomes possible by providing a frame of reference and supporting instrumentation to enable multiple process stages to proceed concurrently. This is possible because datagenous decouples traditional development dependencies in three ways:
- data control flows (schema) are decoupled from system data flows (service invocation) - allowing for zero cost schema iteration and versioning
This decoupling allowed for our clients to model relationships across heterogeneous data structures, normalize common views across these structures, and provide unified and uniform data services across multiple PLMs to client applications, enabling agile development of unifying schema while transparently supporting client application development.
- service invocation is decoupled from service deployment topology - allowing for services to be scoped, centralized, federated or distributed depending on IT environment conditions and performance/scalability requirements and therefore hedging IT risk -
In this particular case our client was able to defer tight coupling of disparate PLM systems and avoid the need for purchasing additional licenses, selecting one system and migrating from one system to another, or consolidate physical data from both systems in one data center, instead being able to use datagenous enabled virtual data center capabilities to feed client analytic applications. More importantly these process happened without requiring production systems to stop during PLM integrations and migrations.
- data sample rates and temporal characteristics are decoupled from systems time allowing for temporal reasoning without the need for fusion or interpolation.
The loose coupling of systems allowed our client to synchronize systems that had different reporting intervals without having to change the operational behaviors of underlying systems and processes.
Comparative approaches often focus on ¨to be architectures¨ that do not uncover or adress the implicit semantics and unknowns of how things work in place, nor fully leverage existing infrastructure to establish viability. Unsurprisingly these approaches take on substantial risk, replacing concrete execution with future promises that will eventually deliver on the strategic desires of leadership. Using datagenous our client applied an MVI approach, utilised and acquired knowledge concurrently, mitigated risks, and rapidly responded to changing conditions. With two experts in six weeks, our client accomplished their goals for a project typically scoped for teams six to eight with durations of nine months to two years. They operationalized the value of acquisitions at speed, and at a fraction of the time and cost typically associated with complex integrations and spin-offs.