A Data River Runs Through IT
- TREATING IT AS A CONTINUOUS, REAL-TIME STEAM IS ONLY PRACTICAL WAY TO HANDLE BIG DATA.
- THE OLD APPROACHES, SUCH AS THE ETL METHOD OF BATCHING AND TRANSFORMING DATA, AND DATA WAREHOUSING ARE FADING AWAY.
“If your system hasn’t been revamped in the past two years, it probably needs to be re-architected now.”
Data these days are so big and fast that, for me, the only practical way to handle them is as a continuous stream, end to end. That, of course, is a major disruption to the way we have been doing things.