« Avec Confluent, nous disposons d’une plateforme de streaming de données rapide à mettre en place, dotée d'une configuration automatisée, et qui ne demande aucun effort de maintenance, ce qui nous permet de nous concentrer sur la logique de traitement. »
Mathieu Despriee
Chief Technology Officer, Sencrop
Founded in 2016, Sencrop has quickly become the go-to provider of real-time weather data for farmers across Europe. In a new era of smart agriculture, farmers have taken their businesses digital and regained their decision-making power with data. Having the right data, at the right time streamlines their daily tasks, reduces unnecessary use of chemical inputs, helps them care for their crops and boosts their productivity.
With a network of over 35,000 Internet of Things (IoT) weather stations across Europe, farmers have access to real-time data on rainfall, temperature, humidity, and wind—all of which help them to better manage their business day to day. They can better understand the risk of temperature and humidity triggering a particular disease or pest in their crops, schedule spraying based on wind speeds, or know which fields to tend to, based on rainfall.
Sencrop started with a simple streaming technology, but as the business grew, the organization needed more features and a broader technical ecosystem.The IT team also needed a data streaming solution equipped with more powerful features, including built-in streaming processing, compacted topics, long-retention topics, data schema management, and topic management. The small engineering and development teams didn't have the resources or skill set to self-manage Kafka and handle any potential incidents. To meet these needs, Sencrop decided to adopt Confluent’s fully managed cloud-native data streaming platform.
Technical Solution
Confluent Cloud provides a seamless, plug-and-play architecture—a “serverless approach” which enables Sencrop’s engineers to focus entirely on clusters and topics instead of endless management and operations work.
The start-up has been able to decouple data ingestion, processing, and storage, which has been critical as data sources and complexity increased. Currently, the data comes from partners—around 20 different forecast models and data—along with all the data from customer IoT devices in the field. Sencrop uses Apache Flink® for stream processing, hosted on Amazon Web Services (AWS), and data flowing through a Confluent cluster. This data pipeline also outputs data to a Databricks data lake to fuel statistical analysis and ML model training.
Business Results
Today, data streamed via the Sencrop platform is accurate to the precise field location and moment so that farmers know if weather is changing and can make important decisions around that information. In Europe, there are strict guidelines about when crop spraying can occur. Confluent helps Sencrop’s farmers reduce their risks and maximize their productivity–and crop yields–with less effort and fewer mistakes. With the recent addition of machine learning (ML) models, IT is better equipped to identify anomalies such as device issues, ensuring that data is complete, accurate, and up to date.
Changing climates, increasing governmental regulation and evolving agronomy techniques—these factors have motivated the team to focus on continually developing new innovations in areas where the need for real-time data is most critical to the business’s success.
Sencrop continues to expand further into European markets as well as other continents, with the goal of continuously improving agronomic predictions with data streaming.
“As our business grows, our architecture and capabilities will also need to grow, and we are confident that Confluent will scale with us, seamlessly.”
Lancez-vous sur Confluent dès aujourd'hui
Pour toute inscription, recevez 400 $ à dépenser sous 30 jours.