Home » Solutions » Business Intelligence and Data Warehousing

GemStone in Business Intelligence and Data Warehousing

GemStone in Business Intelligence and Data Warehousing


GemFire and Data Warehousing

In its original form, data warehousing was meant to store an enterprises variegated data in an efficient way and run deep analytics on that data. These analytics were used for business intelligence or for deriving strategic insight into business decisions. The classic example of data warehousing occurs at large financial institutions: banks offloading the daily trade data, market data, risk data etc at the end day to a data warehouse to run deep analysis and provide a trading strategy for the next day. This style of crunching numbers worked well for awhile until banks started to realize the value of real-time business intelligence. That is, on-the-fly analysis of real-time market data. For banks, this allows a ‘live’ trading strategy that can evolve organically with the ever changing market. However, with the old style of data warehousing, such real-time business intelligence was not possible… until now.

When used in tandem with a data warehouse, GemFire acts as a information broker in-between an enterprises data sources and the data warehouse. In the banking example, GemFire is enabled across the enterprise: Front office trading systems, Middle office risk and reference systems, Back office clearing & settlement systems and buy-side Hedge/Mutual funds. The data fabric that spans these systems provides a holistic view of business operations to the data warehouse and can feed snapshots of pertinent data into the warehouse periodically. The warehouse can perform deep analytics against this federated view of the data and provide information that can be used to update trading strategy throughout the day. Clearly, the result here is better overall risk management. Furthermore, GemFire can provide the same holistic view across geographic locations. This style of unrestricted data management greatly improves one’s live trading strategy, especially in foreign exchange. Finally, updates from the data warehouse are pushed to interested clients via the data fabric across physical and geographic locations, making all updates real-time.


• Captures trade data in real time
• Safe stores trades to persistent store using lazy write behind
• Calculates position as trades and market data stream in
• Updates trader GUIs every 1 second
• Updates Risk System using lazy write behind
• Captures and aggregates market data
• Feeds snapshots of market and position data to the warehouse every 1 minute
• Provide customer, transaction and or product profile information in real-time
• Process flow of events in real-time
• Generate actions/decisions based on fixed rules and profile information