Citimobil, one of the leading cab ordering services, started using Exasol's analytics database. The company made this decision to centralize access to analytics for business users and support real-time decision-making.

​

Initially, Citimobile used a bunch of Open Source technologies including MySQL, ClickHouse, Apache Airflow, and Kafka to store and process data. However, in 2019, the company began to actively expand into the regions and the data volume multiplied, exceeding 30TB. At that point, the data technology stack needed to be strengthened for agility and usability.

​

The analytics database was to be the ultimate data access point for both business users and the systems that would be accessing the data. When selecting a solution, Citimobile focused on the following criteria: performance, SQL compliance, availability of window functions for analytics, compatibility with Tableau and Power BI, availability of user-defined functions (UDF) based on Java, LUA, Python languages. Exasol analytics database proved to be the best in terms of a set of parameters.

    "In our own testing, Exasol outperformed the other solutions in terms of speed and scaling of analytics. The auto-indexes are particularly noteworthy - they are automatically generated based on how the data is used and minimize query optimization effort. Also, using Exasol as part of the data warehouse has helped us reduce the time it takes to prepare reports, despite the growing volume of information. In 2020, we plan to maintain the current business growth rate and expect to increase the amount of data to 100TB," says Konstantin Sevostyanov, Head of Analytics Infrastructure at Citimobile.

​

ATK Consulting Group provided access and helped integrate Exasol into the company's operations. Currently, Citimobile's analytics infrastructure includes an Apache Kafka data bus, a ClickHouse-based staging layer that stores data without pre-processing, and the main layer, an Exasol analytics database with processed data and showcases. It is also planned to store part of the staging layer in Exasol for operational access. This architecture allows maintaining high speed of data retrieval from the storage, completeness and reliability of the stored data, as well as quality process of replenishment of the storage.