Hi Just, This sample script is not de

Hi Just, This sample script is not designed for 24/7 environments. So you right, there is time period between you drop partition and create new partition when cube results will not be right. For any environment when processing window is smaller I would do this script a bit differently - I would process cube structure first, then process data then process index - all in separate batches. Also, for your case you might consider creating and processing new partitions first and then drop partitions. As partition processing is done in one trasaction, they'll not be available till very end, but there will be short time period when data will be duplicate, so this is not a good solution for any production environment. For 24/7 environment I would strongly consider having database just for processing - probably on a separate machine. I would do all my processing/aggregation tasks on that separate machine and just after that I would either synchronize cubes or would do database detach/copy/attach.