The Five Steps to Gaining Control of a DevOps Environment

Source – dbta.com

Companies are committed to delivering higher levels of customer satisfaction for their online services. To create a path to continuous service delivery optimization, you need to start with a review of your current approach and toolset against your business needs.

The first step to getting control of and visibility into your DevOps environment is to collect and instrument everything. But how do you get started, and what are the next steps?

DBTA recently held a webinar featuring Russ Savage, product manager, InfluxData, and Chris Churilo director of product marketing, InfluxData, who discussed a five step process for monitoring a DevOps environment.

Customers continue to face challenges including the limitations of solutions, manual capacity planning, the time it takes to collect new metrics, and an overwhelming variety of solutions, Churilo and Savage explained.

Step 1 is to collect the data:

  • Inventory what you are collecting
  • Why are you collecting – understanding the goal
  • How do you collect
    • Known file locations or specialized agents
    • Push vs Pull metric collection
  • What are you missing?

Step 2 is to correlate and triage:

  • Data dimensionality: different frequencies, different time frames and different sources
  • Established baselines for basic availability
  • What are you missing?

Step 3 is to identify trends:

  • Forecast based on historical data
  • Establish thresholds for alerting
  • Managing inventory – do you know what your safety levels are?

Influxdata can help meet these steps. The platform offers new workload requirements such as high value of real-time writes and retention polices. The platform also offers time-based queries along with scalability and availability.