Menu
A major Middle Eastern mobile operator was facing significant challenges in managing and processing the vast amounts of data generated from its network infrastructure and customer base. With millions of subscribers and a constantly growing demand for data services, the operator struggled to process and act on real-time data, particularly in areas like network monitoring, customer analytics, and billing operations.
Their existing infrastructure was largely batch-based, leading to significant delays in processing critical data such as call records, data usage, customer behavior, and fraud detection. This resulted in:
The operator needed a real-time data streaming solution that was scalable, fault-tolerant, and could integrate with their existing systems. After partnering with Wizuda and evaluating various technologies, the decision was made to implement Apache Kafka as the core platform for processing and distributing real-time data across their network.
Wizuda MFT’s Virtual Servers use key fingerprints to allow clients to connect. This involves verifying and accepting the client’s fingerprint, and storing it, or the key it refers to locally along with some of the other client’s information to allow future connections to be authenticated automatically. Fingerprints are smaller in size, allowing for connections over untrusted channels (e.g. the internet), where a public key would be too large to send for authentication.
Each transfer job can be created to perform a variety of tasks including, but not limited to;
Wizuda’s unique ‘Health Check’ automatically alerts you of file transfer issues as they occur, enabling you to proactively manage them before they impact the business. Scheduled checks monitor host connectivity, authentication, folder and transfer statuses, alerting users of any system or delivery threshold issues. Alerts are issued in real-time when systems are offline or delivery thresholds are broken. Full monitoring logs and reports enable users to locate and fix issues quickly.
By using Kafka for real-time network monitoring, the operator reduced service outage detection times from hours to minutes. This led to faster resolutions of network issues and improved customer satisfaction.
Kafka’s real-time data streaming allowed the operator to update customer usage balances instantaneously, eliminating discrepancies between actual usage and displayed balance. Prepaid customers, in particular, saw significant improvements in their experience, leading to fewer complaints and reduced churn.
The fraud detection system, powered by Kafka, allowed the operator to identify and respond to fraudulent activity in real-time. By reducing fraud detection times, the company saved millions in potential revenue loss and protected its customers from unauthorized activities.
With Kafka’s distributed architecture, the mobile operator was well-prepared for future growth. The platform’s ability to handle high- throughput data streams provided confidence that the company could seamlessly expand into new areas like 5G and IoT without overhauling its data pipeline.
Kafka enabled real-time analytics and dashboards that provided insights across various business units, from marketing to operations. This data-driven approach allowed the operator to make more informed decisions and offer tailored services to its customers.
By implementing Apache Kafka, Wizuda enabled the mobile operator to successfully transform its data architecture to meet the challenges of real-time data processing. Kafka provided the foundation for scalable, real-time data pipelines that improved network reliability, enhanced customer experiences, and enabled proactive fraud detection. The operator was not only able to meet current demands but also positioned itself for future growth with Kafka’s scalable and fault-tolerant architecture.
