Scalable solutions to support business growth
As well as addressing the needs of small clients, Misys Sophis also provides appropriate technology for larger and growing installations that require scalability at the different key points where bottlenecks are usually identified.
The ultimate goal is to provide users with fast access to all types of critical data and calculations along the STP chain and for risk management purposes. We achieve this when solving several key issues:
- Capture high and growing volumes of transactions
- Process the flow of these transactions along a front-to-back office chain
- Have fast read access to huge volumes of instruments and positions data
- Distribute increasing real-time messages of notification and coherency across the whole installation
- Compute large amounts of calculations and above all return results within a short time frame
- Provide access to functionality at different sites rapidly
- Provide load balancing capability to cope with peak activity
- Provide reliable installations with disaster recovery sites
Misys Sophis brings a complete answer to these challenges with an architecture that is both horizontally and vertically scalable. We provide a distributed grid cache and grid computation environment, proven scalable real-time middleware and the facility to set up business continuity plans.
- Each individual server-side component of Misys Sophis' Service Oriented Architecture can be distributed to different servers. This allows organisations to capture and process up to 2 million trades a day
- The underlying Oracle database can be scaled over multiple machines in a RAC configuration
- Tibco EMS middleware is used to handle event and price quotation messaging. It supports high volumes of messages without any risk of bottlenecks. It is stress-tested with over 1,000 Misys Sophis client applications in production and over 12,000 real-time subscriptions. Persistency, stability and reliability are maintained through fault tolerance. Load balancing allows the system to operate at an optimum level even during peak hours and single points of failure are removed
- Grid computing of complex, resource-heavy calculations in short time frames is ensured by the Misys Sophis Calculation Server and its partner Datasynapse. Distributed calculation on grid farms is available for P&L calculation, Monte Carlo algorithms, Greeks, multi entities end of day procedure, simulated scenarios and VaR
- Grid Cache is used to ensure fast access to critical data. Thanks to this technology, which is supported by our partner Gemstone, portfolio loading, transaction aggregation and portfolio calculation times are kept as short as possible, due to effective sharing of resources
