This innovation brings true artificial intelligence to CA Service Virtualization 8.0, a solution that is a critical component in addressing the challenges of DevOps by automating the reasoning, planning, learning and communication associated with the creation of virtual services.
In this application economy, time-to-market, development costs and application quality are significant challenges. CA Service Virtualization accelerates the delivery of high quality software solutions and uses patented technology to eliminate development and testing constraints by virtualising a development system’s behaviour, performance and data so the need for live systems is eliminated or reduced.
The PhD research project conducted by Swinburne University of Technology’s Ms Miao Du has helped bring service virtualisation technology to the next level. As a result of this research, two international peer-reviewed conference papers have been published on opaque data processing and four United States patent applications have been filed.
CA Technologies has worked with more than 40 universities around the world over the last decade and this innovation represents one of the most significant contributions to its products.
“It’s an incredible accolade for Miao and the team to be part of a program that has contributed significantly to a product’s development process and will be sold to customers across the world to help them thrive in the application economy,” said Dr Steve Versteeg, vice president of Research at CA Labs, Melbourne.
Professor John Grundy, Dean, School of Software and Electrical Engineering, Swinburne University of Technology said, “We are very excited with the results of our ARC Linkage project with CA. It is very rewarding to see the commercialisation of our work and to know that organisations worldwide will benefit from our discovery and be able to get applications to market faster with improved software development and testing environments.”
The project was managed jointly by the company’s research arm, CA Labs, and Swinburne as part of an Australian Research Council (ARC) Linkage grant led by Swinburne’s Professor Jun Han. The project combines Swinburne’s world-leading expertise in software engineering with the expertise of CA Technologies in service virtualisation and predictive analysis.
Ms Du’s research was supervised by Associate Professor Dr Jean-Guy Schneider and Professor John Grundy from Swinburne and Steve Versteeg from CA Labs. This team worked closely with CA senior service virtualisation architects in Dallas, Texas to identify new areas to be explored.
About the research
The research used a genome-sequencing algorithm called Needleman-Wunsch to detect byte-level patterns in messages sent between services. This new process, dubbed opaque data processing, enables a service to be virtualised in the absence of expert knowledge, explicit documentation, or needing to know the message structure.
Opaque data processing eliminates the need for the typical Data Protocol Handler (DPH). Instead, it matches requests based on byte-level patterns and provides accurate responses based on a company’ service transaction library. The larger the library, the more accurate the response as the more data it has to learn from, the more precise it becomes.
The technique means that a vastly expanded range of service protocols can be virtualised. For example, consider a legacy mainframe system where the system expert has retired. Opaque data processing can virtualise the service protocols of this system, which was previously impossible.
It also lowers the knowledge barrier for users, eliminating the need for them to understand the full construct of a protocol, WADL or Copybook.
CA Technologies has been running programs in association with Swinburne University of Technology since 2006.