Many trading systems in use today have rigid data models which are difficult to extend, can't keep up with front office innovation, and leave the window open to fraud by using the "cancel then re-book" methodology when making modifications to a trade.

In January 2008 Quartet Financial Systems completed the first implementation of its ActiveStream product that aims to address these weaknesses with a service-oriented architecture approach to trade lifecycle management while utilising existing services (pricing, settlement, corporate actions, trade capture, etc.) of a company.  For example, it can use a company's quantitative pricing library to generate payoffs in middle office (settlements, etc.), and it saves on back office personnel normally used for reconciliation, manual calculations for complex exotics payoffs, confirmations, etc

The company initially saw the main benefit as enabling the front office to go to market quicker with new instrument types, as it allows the back and middle office, with less staff (no need for manual intervention any longer), to maintain those deal types and keep up with trader/quant innovation.

But with innovation in complex financial instruments no longer flavour of the month it is attracting more interest for its unique bi-temporal engine, which clearly separates business and system time lines. By recording the cause of a deal change rather than simply recording the deal states it helps reduce booking errors and potential fraud across all types of financial trades.

It client, a large American bank found that 80% of the work being done by the operations staff was based on this cancel / re-book scenario. Not only was this a huge waste of company resources, but the volume of this manual checking left them wide-open to fraudulent practices - both situations that have now been addressed.

Real-time risk

Also in 2008, Quartet went live with the first customers for its ActivePivot solution that allows real-time aggregation and manipulation of data to support real-time intraday risk calculations. Calyon and Natixis are both using the in-memory, object-based, OLAP and aggregation engine to allow users to "slice and dice" real-time risk data for derivatives within Excel. 

ActivePivot provides greater efficiency by aggregating risk figures (such as VaR scenarios) as they come out of the calculation engine(s) without having to wait for the last figure to be calculated before starting the aggregation. When the last figure comes out of the calculator, the aggregations are ready on the ActivePivot server to be queried by users, thus saving several hours of computational time. It also saves on database storage as it can aggregate objects (such as large simulation vectors) directly from the data source, batch or real time, without having to store them in a relational database. 

When data needs to be refreshed intra-day, either because trades change or market data changes, it can update relevant aggregations without the need to recalculate everything.  This provides a new model where a generic OLAP aggregation tool can be plugged directly to a calculation grid, and provide users with flexible slice and dice reporting. 

Finextra verdict: Two very clever technologies that are addressing some major issues in the industry today - fraud and real-time risk - while maintaining alignment with flexible SOA methodologies and trends in data manipulation for grid/cloud environments.