The cloud is designed for batch processing for statements, regulatory filings, testing of algorithms against historical data and the prior day’s trading data, and other analytics where I/O latency is not crucial.
There are a number of benefits to using the cloud, Jarrod Yuster, CEO at Pico Quantitative Trading and one of the beta testers for the financial cloud, tells El Reg. For one thing, a lot of financial firms get access to the day’s trading data at 11:30 PM local time for the NYSE or Euronext markets, and then begin downloading that data across high-speed (and expensive) wide area network links to their own data centers. This download process takes many hours and lots of dough.
But if you locate the analytical applications that chew on the data inside the NYSE Euronext data centers, you can access the data locally using the Secure Financial Transaction Infrastructure (SFTI) network backbone that the exchange operator uses to link other exchanges, its 1,250 customers, and other market centers into its trading engines and data aggregation services. This backbone is used to create the NYSE Superfeed, which is an aggregation of 170 high-speed exchange and aggregated market data feed handlers. The biggest banks in the world have 80GB/sec pipes (that is bytes, not bits) into the SFTI network, according to Young.„