The architecture of a robust modern Analytics Platform has to address the following needs.
- A cost effective well governed and well protected data acquisition platform
- A data scientist friendly data consumption interface for batch, ad-hoc and stream analytics
- A robust extensible data science toolkit
- A business user friendly data consumption interface for the latest ad-hoc data insight tools with robust visualization
- A robust collaborative decision support enabling workflow and user interaction application framework
- A robust science models for domain specific problem solving
- A fast data scalable and reliable data stream infrastructure
- An extensible stream analytics, monitoring and alerting mechanism
- An extensible semantic context library for data platform with runtime extensibility
- A mature governance, security and privacy model that spans the heterogeneous infrastructure
Profitics has built, tested and deployed various best practice architectures for a wide variety of analytics needs with different levels of emphasis on different aspects of the above mentioned needs.