With autumn’s arrival, farmers are nearing the end of their harvest and planning for next year is well underway. Similarly for enterprise IT leaders, autumn is also the season to plan for the coming year. One of the critical strategic planning questions for 2016 is how to mange the introduction of greenfield ops with brownfield (legacy) ops. [This is the central topic of an upcoming executive webcast on October 13, featuring an expert panel of C-level execs – more on this at the blog’s end.]
Brownfield and Greenfield Defined
‘Brownfield’ and ‘greenfield’ are both software development terms. Brownfield development describes problem spaces needing the development and deployment of new software systems within the immediate presence of existing (legacy) software applications/services. This implies that any new software architecture must take into account and coexist with live software already in production. A brownfield project usually refers to a major upgrade, or a redevelopment of an existing live application where there are issues like backwards compatibility to existing file formats, interfaces, modules, etc. On the other hand, greenfield development lacks any constraints imposed by prior work.
Both terms relate directly to IT production operations, i.e. maintaining where application workloads run and ensuring that they run with the necessary availability and performance. Brownfield ops is IT operating as traditional IT services with legacy tools and processes, emphasizing safety and accuracy – what a traditional IT organization does best. Greenfield emphasizes agility and speed, like a digital startup, and is closely tied to cloud adoption and new tools / processes. See the attributes below that summarize the contrasts between brownfield and greenfield ops.
Brownfield Ops Attributes
- Slow change, but reliable & scalable
- Maintaining workloads across on-premise data centers
- Not cloud platform enabled
- Waterfall release cycles
- Legacy, monolithic tools
- Legacy, serial processes
Greenfield Ops Attributes
- Fast change, innovative and more tolerant of bugs
- Maintaining workloads across externally sourced infrastructures
- Cloud platform enabled
- Agile development cycles
- New, composable tools
- New, collaborative processes
The Executive Dilemma with Brownfield and Greenfield
The ongoing dilemma for enterprise CIOs and CTOs has been determining how aggressively their IT organization should introduce greenfield ops alongside existing brownfield ops. The key capabilities needed to support security, management, seamless deployment choice, ITIL processes and so on, haven’t matured or existed across greenfield and brownfield – creating operational risk. At the same time, ignoring cloud altogether is dangerous; all the controls put in place to safeguard the business are bypassed the first time a developer puts their credit card into Amazon. Similarly, from an application (and business) architecture perspective, SaaS solutions are often fully vertically implemented vs. horizontally integrated, leading to user frustration, inefficiency and potential for disintermediation. Regardless of these concerns, businesses are starting to tell CIOs and CTOs to ‘deploy cloud’ exacerbating the dilemma. It is portrayed as ‘if everyone else is doing it, why aren’t we?’
The cloud has emerged as something of a panacea for digital transformation. Larger enterprises in regulated industries are now starting to (finally) embrace the cloud as well, and are starting to move application workloads to the cloud. However, frustration amongst enterprise IT leaders lingers – from concerns over the cloud’s ability to accommodate truly business critical applications, to questions about governance and compliance issues around business data stored on cloud platforms.
Extensive survey data confirms these frustrations, echoed both in the findings presented by 451 Research at their Hosting & Cloud Transformation Summit earlier this month, as well as the findings collected from nearly 1,600 enterprise IT decision makers in the Cloud Reality Check 2015 report from NTT Communications.
Summary Findings of 2015 State of Cloud Report (source: NTT Communications)
- No definitive answers for which app goes where – instead, a range of variables is in play, including the nature of the application, its maturity, the industry sector in question and even the region the decisions are made in.
- A surprising number of ICT decision makers don’t believe cloud is living up to its potential – but cloud will claim a growing share of ICT budgets over the coming years.
- ICT decision makers are still unconvinced by PaaS – while IaaS is the platform of choice for nearly half of respondents, PaaS deployment trails the rest of the market.
- Some 10 percent of apps will never migrate to cloud – this appears to be the case more so in highly regulated and industrial sectors.
- Cloud rivals corporate data center in terms of its potential to scale in line with demand, as well as its ability to integrate with other areas of the business.
- Cutting through the hype around cloud is difficult – nearly half of the ICT decision makers we polled agreed that managing cloud vendors is confusing and challenging.
- Bimodal IT is hard to do – more time is spent maintaining the current performance of applications than building functionality for the future.
- Service providers need to provide smoother migration paths and greater assistance to complex organizations making the journey to a digital business. Disillusionment has set in and the process of migrating to the cloud is still more trouble than it is worth for many ICT decision makers.
- Security-sensitive apps still aren’t moving to the cloud – nearly a third of ICT decision makers voiced security, compliance and governance as reasons for not migrating to cloud.
Two particular takeaways are worth further discussion here. First is point #1 mentioned above, that one size in cloud doesn’t fit for all applications. Enterprises are confronted with an increasing choice of cloud platforms to deploy their applications (which averages around 100 applications per enterprise in the USA, according to the report). Applications can run on premise in a data center or they can run on top a cloud platform. Of the applications that are running on cloud, the average enterprise is using 4 separate cloud platforms. And this deployment of multi-platform isn’t going to change anytime soon. This increases that need for management tools that can work across platforms – call it a 360- degree view – and helping to pulling it all together into a cohesive whole.
The second notable point is #7, that bimodal IT is hard to do. Bimodal IT been getting recent airplay, a concept introduced by Gartner to advocate organizing and operating brownfield and greenfield separately, bifurcating the IT organization to run at two different speeds. One group is tasked with ‘keeping the lights on’ functions, and the other on ‘business innovation’ functions. But in a recent article in ComputerWorld, CIOs/CTOs are recognizing that it’s not easy to bifurcate the team partly because there’s a risk of maintaining morale for the group that’s focused on traditional unsexy stuff while another group is exclusively focused on the exciting, innovative stuff.
A recent Forbes article goes even further, calling Gartner’s bimodal IT a “recipe for disaster.” The central challenge with bimodal IT, argues Jason Bloomberg of technology advisory firm, Intellyx, is that it encourages IT management to shift their transformation effort away from the slow, “mode 1” IT (traditional IT services), when the very focus should also be to modernize it. Brownfield ops isn’t going away anytime soon and many end-to-end digital transformation projects are the very integration of brownfield and greenfield on each end. The immortality of mainframes in the enterprise is a well-articulated example of this in the article. Bloomberg further argues that CIOs/CTOs must find ways to transform traditional IT (albeit its challenges) as well as meet the end-to-end digital transformation objectives of the enterprise; brownfield and greenfield ops co-existing with value-creating innovative projects across them end-to-end.
What This All Means for Your 2016 Plans and Priorities
What’s the net-net of this all? Enterprise IT is becoming more hybrid, bimodal, multi-platform, (multi-modal!) and difficult. The reality of cloud in 2015 is a complex one – potentially as complex as the native on-premises world it was supposed to replace.
So for 2016, what should your strategy be around brownfield and greenfield ops? In simple terms, do you focus on modernizing the brownfield, or do you proverbial ‘pull the plug’ and try to move it all over to greenfield? How should you implement bimodal IT, or is it really multi-modal IT? And then what does this all mean to the change in your management tools and processes for 2016?
Good answers are starting to emerge. New frameworks can serve as a clarifying guide for implementation. Greenfield principles can be applied to modernize the brownfield as well. The key systems management issues to be addressed are around security, orchestration, systems management, and data management. New management tools and approaches are available that are well suited to multi-modal environments, i.e. to bridge the gap and work across private and public cloud platforms, end-to-end. New tools can now support the growing volume of management data generated in these environments, as well as the dynamic and federated aspects of such hybrid environments – even with a partial and inaccurate CMDB, working across domain silos and getting teams to collaborate better.
Moreover, the thoughtful adoption of greenfield ops principles into your brownfield ops can also serve as a catalyst to reinvent the way you do IT ops end-to-end, that is, around people (organization), process (workflow), and technology (tools). Simply put, best practices around hybrid management and monitoring are available now to help you to better plan and budget for 2016.
Learn More, Download the Webinar “Brown to Green Field Ops: Frameworks for Management & Monitoring”
The pressure for digital transformation is high, yet the suitability of enterprise apps in real world cloud is hybrid, bimodal, multi-platform and difficult. How should you plan for operational management and monitoring as you look to 2016? Panelists:
- Andy Brown, former CTO at Bank of America, UBS, Credit Suisse
- Kalyan Kumar, SVP & CTO – Digital & IT Transformation, HCL Technologies
- Richard Plane, CTO – Cloud & IT Transformation, Cisco Systems
- Phil Tee, CEO & Co-founder, Moogsoft
Get started today with a free trial of Moogsoft AIOps — a next generation approach to IT Operations and Event Management. Driven by real-time data science, Moogsoft AIOps helps IT Operations and Development teams detect anomalies across your production stack of applications, infrastructure and monitoring tools all under a single pane of glass.