Public and Utilities
PGS Software | Part of Xebia
Kynetec is the global leader in animal health and agricultural market insights. The company has a long history of market research expertise, specializing in animal health, animal nutrition, crop protection, farm machinery and equipment, seed/biotech and fertilizers. Kynetec’s number one priority is to deliver the highest-quality insights and foresights at the right time to enable its clients to confidently make the best decisions for their business.
Kynetec was formed after the management buyout of the Animal & Crop Health division of GfK. The company’s employees are located across 30 major agriculture and animal health countries. Its coverage extends to multiple markets, where Kynetec regularly undertakes research projects in over 80 countries.
The research business is a data-rich one. Kynetec processes large amounts of data to create industry reports for its clients around the world. Working on such sizable data volumes, to some extent manually, took up a lot of time for Kynetec’s analysts. Implementing an MVP was the company’s first step towards introducing automation. Later, Kynetec decided to expand this solution to free up its data analysts to work on more meaningful tasks. This is where Xebia stepped in to help.
From MVP to Serverless Storage
Kynetec initially used an MVP with a cloud-managed SQL database service as its main data processing platform. With Xebia’s help, the MVP was transitioned to a file-based storage service using serverless features (where feasible) to minimize the resource footprint when not in use. Raw data can now be “cleaned” significantly faster and more efficiently, enabling analysts to spend more time creating value-added insights.
AWS: A Smart Solution
Thanks to Xebia’s managed Spark computing clusters, Kynetec’s average raw data processing times were reduced by 77%. A team was assembled consisting of three developers and one project manager/Scrum master. Xebia used AWS Cloud as its base: a serverless solution for handling large amounts of data while allowing for the continuous integration of big data. Various AWS technologies coupled with Python and Scala were used as well as tools such as Jenkins, Bitbucket, JIRA, Confluence, Docker, Terraform and Terragrunt. The efficiency of the new processes meant that data analysts could spend more time doing other work.