The in-memory financial risk and retail applications announced today by SAS barely scratch the surface of a big-time trend that is changing the face of analysis and decision support. When speed is the name of the game, the only way to go is to handle the processing in solid-state memory. Thus we're seeing all sorts of announcements that are headed in one direction: marrying analytics with fast hardware architectures.
"Everything is moving away from rotational storage toward solid-state, in-memory processing," observes Forrester Analyst James Kobielus. "SAS is simply responding to the never-ending requirements of its huge customer base for faster analytics, more efficient analytics, more real-time analytics and more flexible execution of predictive models."The in-memory-powered applications announced today by SAS, which are due by the end of the year, are just the latest examples fitting a big trend. SAS was already headed in this direction by way of DIY grid offerings and in-database processing partnerships with data warehousing vendors. Here are a few more proof points:
Long story short, the future will see analytic applications increasingly pre-integrated and pre-tuned to run on state-of-the-art hardware so you can take advantage of brute-force processing power, solid-state memory and massively parallel processing. It's the key to getting fast insight in a world in which complex analyses and big data are increasingly the norm.As today's announcements from SAS suggest, the future will see analytic applications increasingly pre-integrated and pre-tuned to run on state-of-the-art hardware.