 "How Agile Operations can increase business productivity")
By Abhilash Purushothaman In today’s application driven economy, the ability to rapidly deliver high-quality software that transforms Go-to Market has become a major point of differentiation for organizations across industry sectors. As a result, ‘experience loyalty’ has taken precedence over ‘brand loyalty’, meaning if customers (or end- users) are unhappy with the app performance, they will have no qualms in leaving a brand and moving on to the next one. In response, businesses are pursuing digital transformation strategies by adding digital components to all of their products and services that not only enhances the user experience, but also provides alternate route to wider market segments.
To that effect, organizations are increasingly adopting DevOps to transform their businesses. While agile development and continuous delivery remains a top priority, IT operations teams are also aiming to become more agile, without sacrificing the operational stability that has been their claim to fame.
Operations groups have been managing and monitoring infrastructure and applications since the emergence of IT. But, with dynamic evolution of IT infrastructure and emergence of new application-based technologies, this task has become increasingly ubiquitous. Many enterprises are experiencing ‘are we in a maze?’ complex with their decades-old legacy systems. This leads to compromised user experience as well as lack of any visibility or control into business service performance, resulting in extended ‘mean-time-to-resolution’ (MTTR), and ultimately leading to the point of business experiencing an erosion of ‘experience loyalty’.
But what if you could trace transaction movement across cloud, mobile and legacy systems to quickly pinpoint root causes of issues? And what if you could feed that information back into dev/test systems to improve application performance at the code level? And what if you could combine infra & app performance uniquely to eliminate finger pointing, blame games and long resolution cycles?
In my interactions with several CXOs, I hear that as their organizations grew, their IT environments became more of ‘black-box’ with a mix of legacy and modern infrastructure and application components assembled over time. And in many cases, each of these components came with or required its own management systems. They were also faced with many “sources of truth” with disparate tools & silos of technologies, which leads to ‘blame-games’ and finger pointing across various teams. Believe it or not, in the cases when these disparate tools do actually produce valuable data, the onus still falls on some ‘smart dude(s)’ in the Ops team to collate and triage that information before it can result in any kind of meaningful action or resolution.
As businesses get rewritten and business models keep changing, in order to win in the new app economy, many customers are going to go through new digital transformation imperatives. This is why DevOps is important and even more why the “Ops” in DevOps becomes more relevant. Old IT Ops doesn’t work with new infrastructure as it severely lacks speed, scale and collaboration. As organizations move to agile for IT operations, the concern is that Agile doesn’t turn into Fragile. The key question that many IT Ops execs are asking is “How to maintain quality in face of these dynamic changes?”
The business compensates by setting bigger goals, and IT sacrifices manageability and reliability in favor of innovation and speed to meet aggressive deadlines. As a result, more and more technical debt accumulates, increasing infrastructure fragility and management burden—and restarting the destructive cycle anew. We’ve advocated time and again, that in order to overcome these challenges, traditional, fragmented IT organizations must transcend silo boundaries and streamline efforts to collaboratively manage complete services from the customers’ perspective. In other words, they must embrace Agile Operations (Agile Ops) framework. Infrastructure is more dynamic, distributed and complex than ever before. The volume, variety and velocity of changes that IT Operations faces far exceeds the capability of current monitoring tools. A new, more agile monitoring approach is needed.
Agile Operations, on one side, specializes in the deployment, operations and ongoing support of digital services and applications created in agile development and continuous delivery environments. On the other, it emphasizes on the importance of running a strict cadence around collaboration of IT Ops teams with Dev and QA teams to create a 360-degree agile feedback loop resulting in unified monitoring and continuous improvement of the end-user experience across underlying service-delivery infrastructure.
When deployed within a DevOps framework, Agile Operations helps IT Ops teams to quickly pinpoint and resolve performance issues (most notably before they compromise user experience in production landscapes) as well as unify application and infrastructure analytics into a single unified view and enable a more proactive approach to IT Ops management to reduce the impact of technical debt and thereby adapt quickly to ever-changing business goals. These capabilities to transition from ‘legacy IT’ to ‘Agile Ops’ enable the Operations teams to marry the stability they’ve always been known for—with the agility required to compete in the application economy.
Improve scale, speed and quality through collaboration
If we look at the current state, our focus in IT Operations has generally been technology related. We’ve become adept at monitoring large monolithic systems of record, avoiding risk and resisting change. But this must change. IT operations must be capable of supporting new agile, dynamic infrastructure typified by Big Data, API services and cloud, so today’s IT monitoring processes must be capable of ensuring new infrastructure performs reliably at the SCALE they designed for. Also, the processes built around reliability and risk avoidance must change to support the SPEED of delivery never seen before. This no longer means trying to monitor every diagnostic to prevent failure – but discovering problems during an accelerated development and release cycles – even before production. And of course ensuring QUALITY and that all important end-user experience –is critical – so for Agile Operations agility cannot mean Operational fragility.
Given the need to churn out faster releases, most companies don’t yet have the processes and tools to support agile delivery. Development and Operations teams still grapple with ongoing misunderstandings because they neither have the accountability nor have the collaboration framework, to keep track of an application lifecycle across conception, design, development, testing, production and patches. Organizations still have a long way to go in terms of collaboration between Dev and Ops teams and this lack of collaboration has often led to software release / upgrade failure, the inability to scale to meet unforeseen demand, resource exhaustion and configuration errors. As a result of this, Development and Operation teams usually engage in a blame game which affects their scope of work and the organization at large.
It’s high time that CXOs realize the fact that application quality will (only) improve when Operations is active in new product development – like for example establishing comprehensive end-to-end visibility and full unified monitoring during pre-production. Using an agile Ops framework with built-in Dev-Ops collaboration cadence; Ops and Dev are able to share in release sign-offs.
Three mantras to achieving Agile Delivery
The first step towards rectifying this lies in bringing down the invisible wall that has traditionally separated the development and operations teams. In place of this wall, we need to build a bridge which consist of automation and collaborative workflow, the catalyst to agile delivery. This involves establishing a strong pool of tools which will enable both teams to keep a track of an application even in the production stage.
Automation - By automating deployments, one is able to reduce the chance of error when development teams pass code and instructions to operations. In an ideal world, developers can specify the exact components that are needed for the application. The automation process enables all of these requirements to be applied as part of the application’s release package. In the case where an application encounters significant errors, the other key element to automate is a rollback capability. Here, what a team needs is a controlled environment wherein they will be able to revert back to the previous stage. Ideally, this process can be applied from the operations side in a simple series of clicks. Increased visibility - Shared visibility provides the foundation for collaboration once an application has successfully moved into production. This enables a developer to know whether his app is performing or behaving in a different manner in production than it did in the testing environment. Without this insight, the operations team will not be able to detect any unusual application behavior and address the same.
Troubleshooting - This usually emerges as the most common cause of contention between development and operation teams. The key here is that developers need access to production debugging information in a controlled manner, and a troubleshooting toolset so they can keep track the application in the production system. Ideally, with the visibility mentioned above, a developer can actually see why an error occurred, and avoid unnecessary experimentation to try and replicate the error. This means going beyond basic methods which involve operation teams sharing screenshots of the errors and placing a more systematic system which enables the developer to solve the issue.
Sail on Agility
Organizations that embrace agile development while relying on disconnected operations processes are slowly realizing that much of the value gained from agile development is negated. However, given the circumstances, the development approach also needs to take into consideration the operations and delivery side of applications. Once a business starts with these basic elements, they can build and refine their agile delivery even further, thereby scaling up their delivery and meeting customer demands.
The author is Country Director, Enterprise Management Business, CA Technologies, India.
