Designing Real-Time Cubes - Professional SQL Server Analysis Services 2005 with MDX - page 952

Figure 18-10
Proactive Caching Using Timed Updates
The second method of proactive caching is to have the dimension and partition
data updated periodically. While this approach could hardly be considered sophist-
icated, there is no doubt it gets the job done and doesn't take much in the way of
setup. Here is how it works: You set proactive caching to update any new source
data and itself (the cache) at a predetermined time. For example, if you want to
set the update at "24 hours since last process," you set a proactive caching prop-
erty which ensures the MOLAP cache is rebuilt every 24 hours. In the long latency
scenario you would typically not set the latency property since you want the new
data to be available as soon as the MOLAP cache is rebuilt. You need to specify
the option of when to rebuild the cache using the option "Update the cache peri-
odically" as shown in Figure 18-11 . This option ensures that the MOLAP cache is
rebuilt every 24 hours. However you should be aware that the cache update occurs
24 hours after the previous update. For example, on the first day if the processing
started at 12 midnight and it took 30 minutes for the cache to be updated then on
the second day the cache update will start at 12:30am instead of 12 midnight. It