Beyond ONE

Deploying Vertica Eon Mode for scalable and flexible data analytics

The team talked to Beyond ONE’s professional services partner, Integrated Digital Technologies, about a more scalable license structure. Integrated Digital Technologies, with its proven expertise, actively collaborated with the Beyond ONE team, ensuring a solid understanding of the company's challenges and aspirations. They worked together on a detailed analysis of workflows and requirements, defining a customized solution to address Beyond ONE’s unique challenges.

Kurakula commented, “Our partner introduced us to Vertica Eon Mode. This offers the ability to separate compute and storage, as well as support containerization which makes it much easier to scale up when we need it for our resource-heavy queries. It also gave us a pathway to the cloud. This is particularly useful as one of the acquired Virgin Mobile operations (based in Latin America, our first expansion into that market) runs its Vertica environment on AWS, and we are considering a similar move in the Middle East.”

He continued, “We decided to purchase license capacity for 200TB of data which gives us the flexibility to expand without worrying about scalability. We procured the Vertica licenses through Integrated Digital Technologies, and they supported us expertly in setting up the new infrastructure and implementing expanded Vertica use cases.” The customization of Vertica use cases, including those for non-telecommunications departments such as Finance, showcased Integrated Digital Technologies’ commitment to aligning the solution with Beyond ONE’s business objectives.

Moving from an enterprise node-based license model to Vertica Eon Mode also gave the team the opportunity to introduce an Amazon S3 cloud object storage solution as a data repository, with tight integration to an OPTIC data lake, big data solutions, and business intelligence tools to visualize query results. With storage facilities separate from the computing nodes, queries can be executed against much larger data sets without negatively impacting performance at all.

Leveraging advanced analytics within strict data privacy regulations

Beyond ONE operates in a highly regulated market and the team liked Vertica’s InDatabase machine learning capabilities, such as pattern matching and predictive analytics, to ensure that the data is managed and stored according to the necessary regulations and standards. Although the data analytics team is based in United Arab Emirates, the way Vertica manages the data in accordance with strict telecommunications data privacy regulations means that data collected in one country is not moved to another country. Leveraging ODBC or JDBC connections, the team is still able to perform advanced analytics that benefit from a wider data set.

When the Beyond ONE data analytics team began to leverage Vertica’s machine learning capabilities more fully, basing advanced use cases on the functionality, VerticaPy appeared on their radar. VerticaPy is a Python library that is used for machine learning and data science tasks in conjunction with the Vertica database. It allows data scientists and analysts to perform machine learning tasks, such as regression, classification, clustering, and more, directly within the Vertica database. This means they can build and deploy machine learning models on large data sets without needing to extract and preprocess data separately.

Driving data-led strategic decision making with Vertica

The team leverages Vertica for ‘traditional’ telecommunication data analytics use cases, such as churn prediction to improve customer retention, and to determine the most relevant offer considering a customer’s buying behavior. As a result of applying advanced analytics, Beyond ONE’s customer churn rate has significantly reduced.

Kurakula explained the expansion of Vertica beyond these use cases, “Our Vertica data analysis drives strategic decision making throughout our business. Our mobile applications are integrated with Vertica to provide comprehensive usage behavior reports and invoicing to our customers. Our Finance colleagues use it report on revenues against budgets, Product divisions can drill down into the performance of particular models to steer new product development, and Sales colleagues use our analytics extensively to measure the performance of different channels and to calculate commission payments.”