Metrics Platform

Data Ingestion

The Metric Platform supports a wide range of data sources — high-frequency market data, core business databases, and unstructured data such as logs — with plugin-based connectors, CDC, and API integration ensuring minimal-latency ingestion into a unified platform.

Built on DolphinDB's unified storage and computation architecture, the platform eliminates the data redundancy and fragmented stacks of traditional Lambda architectures, establishing a single source of truth across real-time streams and historical batch loads.

Metrics Development

A powerful yet accessible development environment with 2,000+ built-in financial and scientific functions and 10+ specialized engines for aggregation, stateful computation, and real-time scenarios — handling complex cross-asset calculations such as performance attribution and risk exposure with minimal effort.

Metrics are defined in SQL or script with version control and modular architecture, keeping the cost of adding or modifying metrics low. Business analysts iterate independently — no developers required — compressing metric delivery cycles from months to weeks.

Task Scheduling

A built-in scheduling engine spanning the full stream-batch spectrum — from latency-sensitive real-time tasks such as live risk metrics, to daily and monthly batch jobs such as T+1 asset valuation. Open interfaces integrate seamlessly with enterprise schedulers like Control-M, with distributed parallel execution and fault tolerance ensuring reliable delivery within strict time windows.

Metric Management & Publishing

A centralized metric registry maintains consistent definitions, flexible access controls, and full data lineage traceability — meeting compliance, security, and audit requirements. Once configured and tested, metrics publish instantly to the query interface, enabling business users to run multi-dimensional ad hoc analysis without waiting for deployment cycles.

Health Monitoring

Real-time visibility across the full metric lifecycle — from data ingestion and stream table status through computation to result publishing — with resource utilization monitoring throughout. Metric lineage tracing and built-in diagnostics enable rapid root cause analysis and fault resolution. A high-availability cluster with failover and redundancy ensures uninterrupted metric service for critical business decisions.