Report with accuracy and consistency against large changing complex data sets. Create, change and run complex reports on demand as fast as you like.  Run reports at any point in the past (as-at) with no loss in performance.


Cyoda is a distributed platform running across an unlimited number of machines. It breaks up queries and distributes across all available compute nodes, so that reporting performance is linearly scalable. For example, if a report for 1 million trades takes 10 minutes on a 10 machine cluster, then a 100 machine cluster will deliver the same report within a minute. The Cyoda platform can reduce many hours of reporting on a traditional system into minutes or seconds.

As at

Cyoda is a time series store, so updates and changes never overwrite historical data. This means there is no performance loss in reporting at any historical point in time. Traditional systems based on relational databases update stored data and add the previous data to an audit. To run a historical report on a relational database, the system will need to ‘roll-back’ to the required point in time and use the audit log to undo all subsequent updates. Thus the further back the requirement, the longer it takes to deliver. In practice, this means it is only realistic to re-run a report from a very recent point in time.

Ad hoc

Cyoda is flexible and fast, so that reports and queries may be run as and when required.  Users can easily and rapidly change or create new reports against an existing data model.

Powerful indexing

The Cyoda platform uses advanced indexing of the data to enable complex queries to extract subsets of data efficiently.  For example, if Cyoda takes 1 second to extract all 1,000 objects matching a specific query, it would still take the same 1 second, whether the total stored objects is 1000 or 10 million.  This indexing is similar to a mechanism used within relational databases to extract data and avoids the brute force approach of loading huge data sets to find the subset of data required.  Cyoda’s indexing offers a major advantage against a brute approach of querying big data sets.

More Info

Further reading:


STP Processing

Easy Integration

Scale Without Limits

Technical Overview

Data Consistency