Topic: DB2 for LUW
One of the key architectural benefits of BLU Acceleration over classic in-memory databases is that BLU does not require all data to fit in main memory. BLU employs a number of techniques to deal with memory constraints very efficiently, such as late decompression of data, intelligent column-wise prefetching, and smart spilling if needed for extra large joins or aggregations. This session explains these techniques and provides guidelines for how you can monitor and tune the spilling and I/O performance of BLU to obtain best performance on your system.
Click Here to Download
NOTE: These are only open to members of IDUG. If you are not a member, please CLICK HERE for more information.