Newsroom

News about software development, testing and Agile at DVT.
What's the 'software testing' score?

What's the 'software testing' score?

Mario Matthee
Software testing is maturing rapidly in SA, to the point
where we can be very competitive globally,
says DVT's Mario Matthee.

Do you know what is most important to your organisation when it comes to software test automation?


On this note, DVT, in partnership with ITWeb, recently conducted an online Software Testing survey to gain valuable insight into the software testing strategies within South African organisations.


"This software testing survey will not only allow us as a company to understand testing automation within SA, but will also help us understand its maturity and the willingness of companies to outsource locally," says Mario Matthee, head: Global Testing Centre & SQA Professional Services at DVT.


Matthee advises that there are several factors that ought to be taken into consideration when embarking on software testing projects and says that when the methodology is not clearly defined it will impact testing activities negatively as test analysts struggle to define their role and timings.


Dr Jaco van der Merwe, CEO at DVT comments: "As a software development company we often see how companies struggle to release their software with confidence. The accelerated pace at which software updates need to be released makes traditional manual testing often infeasible. We have established a global test centre with a focus on testing automation in Cape Town to address this problem."


Bad business requirement is also cited by Matthee as another factor.


"This is because a test analyst's work is made much easier when high quality business requirements are in place. A negative impact can occur when the importance of testing is still seen as an after-thought even though test teams are in place."


He also believes that test analysts must attend all-important project meetings to improve coverage on risk analysis and business analysts should focus on business value to clients and users, test analysts focus on risk areas.


Another factor that can contribute to the success or failure of a software testing project is lack of communication. According to Matthee, communication is key to the success of any project, especially when it comes to test teams located off-site.


"No regression testing is another factor, with the transition from Waterfall to Agile, most teams struggle to fit regression testing into a two-week sprint. In some cases, teams stop regression testing completely."


Matthee points out that software testing in South Africa is definitely maturing at the bigger organisations, however these organisations started their journey years ago.


"In the SME space, the need for testing has been identified, but cost is a concern," he adds.


Matthee goes on to say that he believes in-house software testing is currently the most popular within South African organisations.


He adds: "the market is definitely shifting towards outsourcing certain aspects of testing, not all of it though. Regression testing is definitely an aspect that can be outsourced and DVT is assisting several clients on starting this journey. Software testing is maturing rapidly in South Africa, to the point where we can be very competitive globally."


The original article was published on ITWeb on 5 November 2015. Click here to read

Related Articles

What is the Databricks
Lakehouse Platform?

The Databricks Lakehouse Platform merges the best elements of data lakes and data warehouses, delivering a solution that offers the reliability and governance of warehouses with the openness and flexibility of data lakes. This unification simplifies your modern data stack, eliminating the barriers in data engineering, analytics, BI, data science, and machine learning. Built on open standards, it maximises flexibility and supports Databricks artificial intelligence. Its unified approach to data management and security streamlines operations and accelerates innovation, crucial in databricks data engineering.

To learn more visit Databricks

What is the Databricks Lakehouse Platform?
Challenges with Data, Analytics & AI

Challenges with
Data, Analytics & AI

Many companies still utilise separate data warehouses for business intelligence (BI) and data lakes for artificial intelligence (AI). Unfortunately, these platforms are not compatible with each other, leading to challenges when trying to achieve successful AI implementation. Currently, most data is stored in data lakes, with a significant portion being unstructured. This unstructured data holds the potential value for AI applications. In some cases, subsets of this data are copied to the data warehouse and organised into structured tables. However, this process of duplicating and storing data in both systems leads to data drift, as changes made in one system may not be reflected in the other. Additionally, separate governance processes are required for each platform, leading to inconsistent control. The different tool stacks used on these platforms further hinder collaboration between support teams. These complexities and inefficiencies result in failed AI efforts, with wasted time, resources, and slow decision-making processes, leaving companies lagging behind their competitors.

Challenges with
Data, Analytics & AI

Many companies still utilise separate data warehouses for business intelligence (BI) and data lakes for artificial intelligence (AI). Unfortunately, these platforms are not compatible with each other, leading to challenges when trying to achieve successful AI implementation. Currently, most data is stored in data lakes, with a significant portion being unstructured. This unstructured data holds the potential value for AI applications. In some cases, subsets of this data are copied to the data warehouse and organised into structured tables. However, this process of duplicating and storing data in both systems leads to data drift, as changes made in one system may not be reflected in the other. Additionally, separate governance processes are required for each platform, leading to inconsistent control. The different tool stacks used on these platforms further hinder collaboration between support teams. These complexities and inefficiencies result in failed AI efforts, with wasted time, resources, and slow decision-making processes, leaving companies lagging behind their competitors.
Challenges with Data, Analytics & AI

Moving forward with a
lakehouse architecture

The data Lakehouse represents a transformative approach to data management, effectively supporting AI and BI applications directly on extensive data lake volumes. This innovative architecture combines the best features of data warehouses and data lakes, enabling robust AI and BI capabilities on large-scale enterprise data. It offers enhanced SQL capabilities, performance features like indexing, and MPP processing for accelerated BI on data lakes. Additionally, the Lakehouse model supports direct file access and is compatible with Python, data science, and AI frameworks, obviating the need for separate data warehouses. By adopting this architecture, organisations benefit from Databricks data engineering and Databricks artificial intelligence, enhancing data quality management. In summary, adopting this lakehouse architecture through Databricks services merges the advantages of data warehouses and lakes, offering a streamlined approach to handling extensive data sets and optimising data quality and management.
Moving forward with a lakehouse architecture