DevOps: Ground-up Approach to Service Management
DevOps (development operations) has become one of the most heard IT practices in the recent times. For decades the IT industry has been delivering code and managing operations successfully, so why this sudden shift in paradigm?
The crux of this revolution is to infuse a culture of continuous service enhancement and show the numbers in the KPI and SLA reports. DevOps teams indulge in client satisfaction surveys, leadership engagement, competitive benchmarking, and dedicated campaigns to generate, qualify, and deploy end-user ideas for improved service.
It is ideal for client to expect improvement in quality and performance, en route to service maturity. It includes enforcing software configuration policy to application build automation, from change control to release governance, etc. Increasing service quality and performance requires a significant effort to analyze and identify the areas for service improvement; this is where the DevOps team joins forces.
Let’s analyze and identify the gaps in development and operations to understand this concept better. We will use the most recommended method:
1. Capture and analyze the following DevOps KPIs to identify scope for service improvement:
- Increase in use of change and release management process
- Reduction in failed changes
- Configuration item (CI) identification and versioning
- Increase in percentage of CI audit
- Incidents (in percentage) due to inaccurate CI information
- Reduction in application build and deployment time
2. Analyze month-over-month metrics and generate a trends report for the SCM, Build, Change, and Release management functions and restructure service levels for continuous service improvement. This trend report highlights any continuous or intermittent breaches in the service. For example, if application builds are taking longer than mandated or production releases are facing issues, invoking emergency change process.
a) In case of a breach, the DevOps team analyzes the cause and act on a mitigation plan.
b) In case there are no breaches the team can work on improving existing performance by using the continuous service improvement (CSI) framework and raise the bar.
3. Track the restructured service levels within the team for verification and feasibility check, for at least one quarter.
4. Once the new service levels are achieved, formally communicate the enhanced and restructured service levels to the end client.
DevOps disrupts the traditional silos of development lifecycle to expedite full-lifecycle implementations and deliver value to client.
Arnab Das is Head of the DevSecOps Practice at Atos Syntel. He has 15 years of experience of leading multiple DevSecOps automation implementations and delivering innovative solutions to solve client business challenges. Arnab is dedicated to formulating and articulating a change roadmap to transform product engineering for Fortune 500 companies.