How BLU Acceleration Really Works
August 19, 2013 1 Comment
Guy Lohman – IBM Master Inventor, Manager of Disruptive Information Management Architectures
Want to know how BLU Acceleration REALLY works? Written by the researchers and developers that invented it, this highly technical overview is not really a white paper, but in fact a contribution to academic literature that was refereed by world experts and presented at the 39th International Conference on Very Large Data Bases (VLDB 2013) in Riva del Garda, Trento, Italy on 27 August 2013.
If you aren’t satisfied with marketing hype and white papers, the details found in this paper should satisfy your curiosity about how BLU Acceleration achieves such dramatic speed-ups and compression without any of the usual “tuning” needed.
Learn how data is compressed, stored in memory and on disk, and scanned during query processing. See how BLU Acceleration’s patented Frequency Partitioning allows predicate application, joins and grouping all to be performed on the highly-compressed values, permitting increased parallelism and improved utilization of all machine resources — CPU, cache, memory, and disk! Find out how the careful engineering behind BLU Acceleration avoids bottlenecks such as latching and exploits modern multi-core processors to maximize the utilization of all your resources; thereby helping your organisation to get answers to complex queries “at the speed of thought.”
Still curious about BLU Acceleration? Find more resources.