Speaker

Mohanpriya having around 7 years of Experience in Performance Testing & Operational Acceptance Testing and currently working as Senior Associate in Cognizant. Worked in various domains such as RETAIL, Manufacturing, Banking domains. Have worked in various applications such as JAVA, ORACLE, .NET, Ecomm, SAP, POS & Sterling OMS applications and Expertise in different performance testing tools such as LoadRunner, JMeter, Neoload & Gatling and Monitoring tools such as Splunk, New Relic, BMC, HP SiteScope, vROps. Possess Good knowledge on Kubernetes & SRE concepts.

Expertise in Mobile Performance Testing, Have been involved in Automation initiatives and developed a JMeter based CI-CD Solution using Gitlab, Automated Test Result Collation & Server Metrics from Monitoring tools. Performed various chaos experiments on Cloud components

Involved in Mentoring the Freshers & guiding them to fulfilling their project requirements. Have been involved in the Recruitment team to identify the right candidates. Supporting the juniors folks in terms of any technical support as well as in contributing for the Transformation / Process Improvements

Title: Test your Infra Resilience in a Boring way or in “The Best way”

Huge amount of effort is required to test your Infrastructure resilience.

 

Is it possible to shrink the timeline of your testing as well as the resource costing?

 

Traditional way of Resilience Testing requires a lot of resources and support from the respective team, when so many resources are required to perform the test, it increases the cost & dependencies. We all know how important timelines are when it comes to the deliverables. We should always have to coordinate with the respective teams to be in sync. otherwise, our ship will drown. In other words, testing might not give you the best results in a stipulated time.

How it's being tested all these days -traditional Way

 

We will prepare the test plan, check for resources availability, explain the testing procedures to them & assign them the tasks to perform the execution as per the test plans. Finally, we will do our analysis. The test completion purely depends on the availability of the resources and if one person is not available it will completely block our pipelined actions, thus affecting the timelines & deliverables.

 

Have you ever thought of trying it with open source where we have access to the environment?

 

Yes, our solution is to achieve it without any dependency on the resources, save some cost, reduce the time & effort.

 

One time Installation and configuration of the agents or the driver on the infrastructure i.e., Hosts, Kubernetes, Containers, Bare metals is required. Execute the test cases using the commands or using dashboards & get the report for the conducted test using any monitoring tool. Here you can add the flavor of Chaos engineering to your infrastructure only if required. Most of us go with a system resilience test with the system or service shutdown. But what if our system crashes because of CPU, Disk, Network. Using this framework, we can inject a variety of scenarios with a combination of attacks & prove the resiliency without any support.