Stateful HTAP (Hybrid Transactional and Analytical Platform)
Cyoda was built to fulfil the technical requirements to implement virtually any organisation’s data processing, batch or real-time and analytics/reporting, with transactional consistency, but without restrictions. A horizontally and elastically scalable platform without limitations to cut the total cost of ownership, implementation, running and change.
What is it not?
- It’s not built on or with Hadoop.
- It’s not an appropriate technology for Ultra-low latency operations such as high-frequency trading.
Distributed, scalable, reliable
- No single point of failure, no bottlenecks
- Dynamically scalable to meet fluctuating demands
- ACID consistency through distributed transactions
- Processing dependency
- Distributed, scalable
- Maintains sequence at the object level
- Write only store
- Object persistence, read and write is taken care of so you model only at the object level.
- STP in user-defined event-driven BPM (Business Process manager) workflow via state-machine (Rules Engine)
- Real-Time processing
- Scales linearly to allow unlimited volumes of data to be processed
- Workflow can be defined by a non-technical user.
- Transparency of the system
- Lower cost of change
Schema agnostic, Schema on read
- No requirement for unified data model
- Leave data in their original format
- Reduce data duplication
- Reduce implementation time
- Model data combining different complex data models and version how and when you want.
- OQL (Object Query for defining) for complex queries and aggregations
- Distributed report load for linear scalability to run reports as fast as required
- Transactional consistency for accurate ad-hoc reporting on live changing data
- As-at and bi-temporal reporting.
- Composite range indexing for constant time data extraction regardless of volumes stored (not brute force reporting)
- Can index raw data as well as secondary generated data
- Enables fast, ad-hoc, flexible reporting, both current and as-at, against arbitrarily complex data and queries. OQL implementation (Object Query Language)
- Combine multiple complex data models and version into a single combined report, define the query and aggregation on the virtual model.