Bringing Agile to the Whole Organization
As McKinsey writes the primary reason for adopting new 'DevOps' practices is to speed your overall organizational innovation activities.
New marketing features for your e-commerce site that boost sales by 5% might require six months of software engineering to bring live via your current methods, but only weeks after you move to new agile, Cloud-centric approaches.
The 'Cloud Native' industry trend of microservices, containers et al, is seeing vendors and standard innovation that very effectively supports this goal. For example Red Hat announced an Openshift-based development environment intended to accelerate this end-to-end lifecycle.
Organizations like BNY Mellon are adopting this platform approach, in their case Cloud Foundry, to enable a new agile culture, and the Harvard article 'Bringing Agile to the Whole Organization' starts to describe the broader, overall organizational transformation that this adoption can be just one component part of.
Technology experts naturally focus on the products and tools, whereas the business gurus look at the whole team involved in this product creation process, from marketing through sales, billing and more. All are involved in the lifecycle, and in a series of articles like The New New Product Development Game and Getting the Most Out of Your Product Development Process they describe how important it is for organizations to apply the best process models to their entire product innovation, not just the software engineering part.
Given the digital world we now live in these principles are relevant to all organizations, a principle very effectively described by How the Software Industry Redefines Product Management, but they are especially so to businesses like SaaS providers or Telcos, because in their case their product is software.
Indeed this is exemplified by the Telco industry, where their traditional industry has been eaten into by 'OTT' firms (Over The Top), refering to Internet-based providers eating into core revenue areas like VoIP. Given the poster child of these new innovative software practices is Netflix, it's an entirely valid statement to consider therefore that software engineering and technology strategy is now a skill set that can yield massive competitve performance advantage.
Network DevOps - Unified Agile Operations
An example of this scenario and the challenges for these more traditional organizations is demonstrated through the NTT Communications case study. In their presentation to the TMF 2016 conference the Transformation From Legacy Operation to Agile Operations, they describe how:
- They have adopted the popular enterprise Cloud Native toolchain combos: Github, Docker, Puppet, Ansible, .. et al
- This is leveraged to automate core platform update methods, such as deploying a new NMS across multiple, global data centres.
- Github is used to store network configurations, and utilized as a base for Configuration Management-based automation, where a new Pull Request is initiated for each Work Order and it caters for core functionality such as Audit Trail.
- Jenkins is used for Continuous Deployment.
- Containers provide for portability across multiple platforms.
Business Transformation challenges
Overall their headline message is one of 'choosing the right tool for the right job', that there is no one particular app that's any more important than another, rather it's their combination applied in the right permutations to automate workflow that's important. NTT also make a critical observation that define the limitations of their progress and with this the need for broader organizational transformation:
On slide 23 showing their future plans and challenges, they highlight how they reached the celiing limits of what toolchains alone can achieve:
- "Each section has its own Agile environment.
- Difficult to promote collaboration over the sections."
In other words tools can only solve tool-related problems - Those created by organizational boundaries et al therefore need organizational level solutions.
Virtualized PLM: Product innovation for software-driven organizations
For other telcos seeking to emulate these agile operations environments the TMF industry forum is particularly well placed to provide business transformation support.
In particular the TMF guide 'IG1129' - Comparison of Traditional vs Virtualized PLM, to highlight because it reflects the most important strategic insight, that DevOps is as much about improving product innovation as it is software development. Indeed for software-driven businesses, it is product development.
Adoption of new NFV platforms will require transformation of Telco organizations at multiple levels: From technology upwards through the sales and marketing organizations. Therefore ideally service providers need one overall metaphor for their entire strategy, and this guide caters for this very effectively.
The opening line defines the core NFVaaS business challenge:
“As operators and enterprises evaluate the introduction of virtualization to network functions they are discovering that the biggest challenges lie in the existing organization, processes and systems.”
The first main paragraph then also provides an effective synopsis of the key market and platform dynamic:
“Introduction of virtualization technologies move the focus of Product Lifecycle Management away from conventional well-defined hardware based services towards the new future concept when services are defined by software blocks and provided using the elements of multiple partner companies. Service components are now represented by cloud-based units which could be rapidly combined to develop a product.”
It then goes on to describe a framework for Virtualized PLM, encompassing:
- A Service Lifecycle defined by User Stories of Concept, Design, Deploy, Operate, Maintain, Retire.
- The 'Future Mode of Operations' set of lifecycles required for administering new virtualized services, from their instantiation for client requirements and subsequent monitoring, auto-scaling etc., there is also regular software upgrades, licencing changes and other activities that make up it's cradle to grave lifetime.
- A core technology platform capable of implementing them, one that achieves real-time, end-to-end provisioning across multiple domains of Cloud, IT and legacy Network resources, achieved through standardized API interfaces and complex 'service chaining' product model configurations.