The concept of bimodal IT, like the game of “Telephone,” has been obscured as people have put their own spin on it, leading to backlash and confusion. Some organizations have moved ahead and implemented continuous delivery models, while others are still transitioning and experiencing growing pains.
But as with any kind of transition, there’s no guarantee of success – merely a set of recommendations for the journey. Nobody said transitioning to a more dynamic and continuous process would be easy. However, failure, fear and skepticism should not give people license to remain stuck in their legacy systems or rush headlong into change. Let’s examine what this term “bimodal IT” actually means, why it makes sense in some cases and how to ease the pain of transition.
Defining Our Terms
As most readers will know, Gartner (News - Alert), Inc. coined the phrase “bimodal IT” in 2014. The analyst firm defined this process within IT infrastructure and operations as “the practice of managing two separate but coherent styles of work: one focused on predictability, the other on exploration.” Mode 1 focuses on predictability and has a goal of stability. It is best used where requirements are well understood in advance, and can be identified by a process of analysis. It includes the necessary investment in renovating and opening up the legacy environment. Mode 2 is exploratory. In this case, the requirements are not well understood in advance. Mode 2 is best suited for areas where an organization cannot make an accurate, detailed, predefined plan because not enough is known about the area. Mode 2 efforts don't presume to predict the future but allow the future to reveal itself in small pieces. Gartner ended its original, short definition with, “Both play an essential role in the digital transformation.”
Bimodal IT creates two separate groups that work at different speeds on segregated systems. It is typically characterized by a waterfall vs. Agile (News - Alert) scenario. Waterfall methodology follows linear, sequential development with distinct goals for each phase. By contrast, Agile processes seek to help teams respond to unpredictability through incremental, iterative work cadences and ongoing feedback. This two-speed method may be the current de facto way of doing things and may remain this way for a while, but slowly and steadily, IT is undergoing massive and fundamental transformation to address customer and enterprise needs for agility.
Challenges and Choices
Not so long ago, the development cycle took six months to a year. Today, it’s not uncommon to see weekly or biweekly releases. What accounts for such a radical change in delivery speed? There are five over-arching trends contributing to IT transformation:
This acceleration of development causes a strain on an organization’s underlying infrastructure. It poses new challenges to IT Operations teams. It requires teams to manage unparalleled amounts of data while predicting and preventing outages, in real time, and maintaining and delivering agile, reliable applications. This increased complexity makes some organizations fearful about transitioning from Mode 1 to Mode 2 completely, as concerns over new processes and operational complexity loom. In order to ensure availability, reliability, performance and security of applications in today’s digital, virtualized and hybrid-cloud environments, new approaches must be employed to provide operational intelligence to ease the transition from Mode 1 to Mode 2.
Easing the Transition
Underlying Gartner’s discussion of bimodal IT is the idea of providing breathing space so that organizations can transform and innovate without crashing and burning. The reason that Agile was created, for instance, was to enable a faster, more responsive process than waterfall practices can offer. However, switching to continuous delivery and integration mode too quickly could prove disastrous for certain systems, as some change carries more inherent risk than others. Following are three best practices to help ease the transition and ensure that applications continue to run at optimal levels.
Applications, whether in Mode 1 or Mode 2, are running on complex and dynamic infrastructures more than ever, with underlying resources constantly changing to meet these applications’ performance requirements. You need visibility into all your data—including performance data, logs and topology—and the ability to visualize all layers of your application infrastructure stack in one place at any point of time. This allows you to identify the root cause of an outage or performance degradation in the past or the present. These tools can also provide the capability to understand the impact of a software release on the operations in the Continuous Delivery and Integration mode (Mode 2). In the absence of such tools, conducting definitive post-mortem analysis is a costly, manual and confusing process – if it can be done at all.
As IT is going through the transformation, it is becoming increasing complex and dynamic. There is a big data problem brewing in IT. Relying solely on traditional IT monitoring tools that trigger numerous alarms makes the job of IT operations teams even more difficult. Understanding all the raw data to make intelligent decisions in real time and sifting through the sea of alarms and telemetry data at the same time poses major challenge to IT operations teams. AI—especially machine learning—is well suited to take all the data and generate the necessary operational intelligence to distinguish critical, service-impacting events from false positives that do not require the immediate attention of an operator. As IT transitions, you need IT operations intelligence that can handle both modes of operations.
The key to preventing outages is to predict issues before they become problems. A twin problem to the one above is that traditional monitoring tools trigger alerts only after a problem has already occurred. Look for solutions that incorporate predictive analytics to alert you to anomalous trends or potentially dangerous issues before they impact your application.
To recap, automated solutions that analyze and provide insight into ever-changing applications and infrastructure topologies are essential to ease and manage this transition. Equipping users with the ability to replay and analyze past incidents and to pinpoint performance degradation root cause, while cutting out the noise and preventing future costly outages and downtime, is important to facilitate the transition. This operational intelligence connects enterprise DevOps and TechOps teams, giving them what they need to quickly address issues as they arise.
Enabling Change
Digital transformation is not "one size fits all.” Each organization needs to carefully assess what can be accelerated and what may need to be kept as-is for now. The phenomenon of bimodal IT will exist for some time as IT continues its journey of transformation. So then, bimodal is not a perpetual phenomenon but rather a step in the direction toward uni-modal IT that is continuous, dynamic and agile. IT operations analytics have the potential to play a major role in the success of changing from Mode 1 to Mode 2, offering visibility into systems and activities and helping to ensure that the business continues to function without interruption.
About the Author:
Dr. Akhil Sahai is an accomplished management and technology leader with 25+ years of experience at large enterprises and at startups. Akhil comes to Perspica from HP Enterprise where, as Sr. Director of Product Management, he envisaged, planned and managed the Solutions Program. At Dell (News - Alert), as Director of Products, Akhil led Product Strategy and Management of Dell’s Converged Infrastructure product line. He also led Gale Technologies, as VP of Products, to its successful acquisition by Dell. Prior to that, at Cisco (News - Alert) he undertook business development for VCE Coalition, and at VMware, he managed global product strategy and management for vCloud Software with focus on applications, and Virtual Appliances product line. He has published 80+ peer-reviewed articles, authored a book, edited another, and chaired multiple International IEEE (News - Alert)/IFIP Conferences. He has filed 20 technology Patents (with 16 granted). He has a Ph.D. from INRIA France and an MBA from Wharton School.