Your data never leaves. The insight does.
AcuBi is a microservice platform that stores only metadata, runs every query at the source, and deploys on your own premises or cloud — engineered for businesses that cannot hand their data to a third party.
Metadata only, by design
- AcuBi
- Metadata repository
- Schema
AcuBi maintains a metadata repository — the structure of your data — and nothing else. When you create a project, the engine pulls only the schema to build the model. When you explore, the platform generates the query, runs it at the source database, and renders the result.
No data is copied, cached or held inside AcuBi. For regulated operators and data-sensitive enterprises, that is the difference between a tool you can adopt and one you cannot.
How the platform is built
- API gateway
- Microservices
- Version control
- Authentication
- An API gateway fronting a set of microservices: configuration (backed by version control), authentication and authorisation, user interface, data rendering, scheduled jobs, and service registry and discovery.
- An application database for the platform's own state, separate from your data.
- Customer-specific microservices added where a business needs them.
- A data visualisation web client that talks to the gateway.
- An intermediate data layer, fed by scheduled jobs, for transforming business-critical KPIs; and direct on-demand access to external data sources for analytics.
The data flow, step by step
- Schema
- AcuBi
Create a project. Select a database, or create a new connection. Choose the schema, then the tables, then the project's privacy setting, and submit — AcuBi reads only the structure and builds the model. To analyse, update the model, explore with the REST client, which generates SQL from your choices, run it at source, and render the visualisation. Every step is repeatable and every query is visible.
Nineteen database dialects and counting
Amazon Redshift, Google BigQuery, Snowflake, MySQL, PostgreSQL, Oracle, Teradata, Microsoft SQL Server, Vertica, Spark, Presto, Impala, Hive, Amazon Athena, Amazon Aurora, MariaDB, Druid, Google Cloud Spanner and MemSQL — with further dialects added to customer requirements.
Security and access
- Filters
- Authentication
User-based data access filters apply inside every report, so one shared dashboard can serve many users with individual privacy. Authentication runs through a dedicated authorisation service. The platform deploys on premises or in your cloud according to your requirements, and Infinetix's own security practice, SmartSec, is built on protecting exactly this kind of estate.
Scale and customisation
Microservices scale independently, scheduled jobs take heavy KPI computation off the query path, and customer-based services extend the platform without forking it. Where configuration is not enough, Infinetix provides custom-made services against your business requirements.