Dave Snowden, Director, Cynefin Centre, Chief Scientific Officer, Cognitive Edge, UK

Eyal Yovel, Senior Solutions Architect, Sauce Labs, UK

Best practice for using Java testing frameworks for testing web apps as well as native apps and running on multiple browsers, multiple browser versions, different OS and different devices.

Many companies build their automated frameworks based on Java testing frameworks such as Junit4, Junit5 and TestNG. They utilize these frameworks for testing their web apps and/or native apps before production in order to improve their quality.

However, simply setting up the automated framework doesn’t guarantee high-quality apps in production. In order to enjoy the full benefit of automated testing, there are additional steps and best practices to consider.

Improving the product quality, decreasing the time to market and receiving quick feedback are key imperatives that need to be supported by any automated framework. To successfully deliver all of this, the framework must be able to run against multiple browsers, multiple browser versions, different OSs (Windows, Mac, Android, iOS) and on different devices.

During this presentation, I will explain:

• the differences between JUnit4, JUnit5 and TestNG
• how to run automated tests in parallel on each of these frameworks
• the best practices for passing parameters such as OS, browsers, device names and versions to the framework
• when to use Data Providers or TestNG XML files, or maybe the best is to use external files such as JSON and CSV files?

In this session, I will also be sharing code examples and multiple tips and tricks for using Java testing frameworks.

Stuart Clark, Senior Technical Leader & Senior Developer Advocate, Cisco Systems, UK

If your network team isn’t among a majority of network developers using Continuous Integration, why not? The concept of Network as Code is based on the idea of storing all network configurations in a Version Control System (VCS) that manages and tracks changes in the network. See how a system can store all network configurations and act as the Single Source of Truth for the configurations. The goal will be to create a complete environment that demonstrates the benefits across the whole network.

In this session, we will look at how network configuration changes are proposed in code branches, just like software and when the changes are ready, the development branch will go through an approval process before it is merged with the main branch and production ready. You will get to know the real benefits of NetDevOps and CI/CD (Continuous Integration and Delivery) and how a build server can automatically deploy and test the proposed configurations in testing, staging and production environments.

View Stuart Clark’s video

Moderator:

• Richard Self, Senior Lecturer in the Governance of Advanced and Emerging Technologies, University of Derby

Panellist:

• Dr Nandakumar Ramakrishnan, Product Owner – DevOps Platform Testing & Chaos Engineering, ING Belgium
• Gil Zilberfeld, CTO, TestinGil, Israel
• Mudit Singh, Director – Marketing and Growth, LambdaTest

Mahitosh Patankar, Head of Engineering, British Telecom, India

Harvard called the Data Scientist as the Sexiest Job of 21st Century. Organizations can have profound impact by having the ability to transform a sea of data into actionable insights.

By mixing math, computers, analytics and topping it with sound business understanding, Data Scientists unravel mysteries and help make decisions backed by data and insights derived by Machine Learning(ML) models.

All the efforts of data scientist in designing , testing and enhancing the algorithms are evaluated when they actually get used, so they need to be continuously tested and improved.

DevOps has bought developers and IT operations together fostering collaboration helping practices like continuous delivery and continuous testing. Data scientists and data science teams face similar challenges of monitoring and modifying ML models, which DevOps concepts can help address.

Data science practice has some notable differences from software development which needs to be considered. They may be amazing at designing and training a model, but making it continuously test, modify, deploy and making it production ready generally becomes difficult and not their usual expertise.

In this session we can see how DevOps best practices can be applied to Data Science making it MLOps (while appreciating the difference) to bring together various actors and process creating ML infused DevOps software delivery process to strengthen, accelerate continuously delivery of insights and values to business.

View Mahitosh Patankar’s video

Gil Zilberfeld, CTO, TestinGil, Israel

This is a case-study of building a quality dashboard with metrics and reports that matter for an application with hundreds of APIs, and multiple front-ends. We collected information from multiple sources and added some “brains” to show the analysed metrics, in terms of covered and uncovered test cases, test quality and more.

Paul Grossman, Lead QA Engineer, Guaranteed Rate, USA

The biggest drawback of test automation is test script maintenance. In this session we will go beyond the Page Object Model to explore tips on updating xPath and CSS locators on the fly. This helps when large amounts of elements are impacted by changes in properties, names or classes, specifically when your boss expects results in a few couple hours. We will also look at the advantages of writing scripts in Plain English, well beyond the rigid GIVEN, WHEN, THEN structure of Cucumber and it’s Step Definition abstraction layer.

Moderator:

• Tom Hoyland, Senior Agile and DevOps Lead, Sky Betting and Gaming, UK

Panellist:

• Paul Grossman, Lead QA Engineer, Guaranteed Rate, USA
• Mudit Singh, Director – Marketing and Growth, LambdaTest
• Eyal Yovel, Senior Solutions Architect, Sauce Labs, Uk
• Victoria Alexiou, Product Manager, Peoplecert, Intl. Greece