In the current scenario that most Telcos find themselves in, there is a strong need to roll out new products faster and in a very stable manner, while the declining revenue growth also demands a cut in IT spend. To make the situation trickier, organizations are structured in siloed team formations where we have a team for IT Development and one for operations. This invariably results in ownership conflicts while handling critical customer-impacting business processes. In such a case, there is a strong ask from most telecom operators to adopt more sustainable and optimized ways of working, with clear and contained ownership within one domain, to enable change while maintaining stability - DevOps appears to be the unanimous first choice.
The solution appears simple at first glance. However, since this is a huge disruption, going with a big bang where we introduce major changes in one shot could lead to failure as teams may take longer time to grasp and adopt all of these changes. Also, there are scenarios where the team’s maturity doesn’t complement the big changes. In order to prevent chaos and ensure DevOps doesn’t result in burden or failure, we suggest a phased approach to move toward embracing effective DevOps.
Initial Set Up
We have laid down these basic checks before we start ideating DevOps for an organization.
Assessing the maturity of the current landscape to see which stage we are on in terms of…
- Ways of Working – With focus on faster roll-out of products to the customer, we recommend an organization adopt a more flexible and iterative delivery methodology such as Agile, so that the benefits of DevOps can be reaped.
- ITSM tooling – The number of processes enabled in it. Basic processes include Incident Management, Change Management, Problem Management, Service Request management.
- Knowledge Management- One of the most underrated yet crucial factors for the progress of any organization is having a good knowledge repository in place. A centralized and up-to-date repository will provide a base for finding scope to automate the repeated procedures and steps, eventually reducing manual efforts and optimizing opex.
- Test Regression suites – Having a good and reliable quality gate in place saves lots of time and effort spent on fixing a cascaded issue realized later in the product lifecycle (incidents). Ensuring good coverage, regression test suite is defined as the initial step to ensure your organization spends lesser time on fixing the “Ops” side of problems once the product is rolled out to production. Hence, the quality gateway has to be well defined and automated as much as possible. Automated regression suites can be the first automation place in the product lifecycle before the product rolls out to production.
- Basic Monitoring and Dashboards – This is indeed a very organization-friendly feature and has a multitude of benefits. Not only does it save a lot of time and effort spent in looking into every application, log and event by providing an updated view on these in one screen, it also makes visualization of the entire product lifecycle easy and user-friendly.
If implemented smartly, this will play a critical role in reducing a large amount of manual Operations work. Additionally, it will aid in the generation of insights for improvement in both the IT domain as well as the business.
On the basis of these parameters, we have listed a few criteria and also defined the maturity levels which we usually see in any organization. The maturity levels in yellow tiles are the basic levels an organization must be at, in order to move toward embracing DevOps.