Database Reference
In-Depth Information
3. To ensure all modified data is stored to disk use SysInternals' Sync Utility tool
(You can download it from http://technet.microsoft.com/en-gb/
sysinternals/bb897 438.aspx ).
By using the Sync Utility tool you can quickly flush
the LUN from the SAN after Analysis Services is
stopped, and before dismounting the LUN.
4.
Based on the SAN vendor, using the command-prompt utility, generate a script for a
snapshot of Analysis Services data folder.
5.
Restart the Analysis Services on the processing server.
For better results on volume,
always mount the disks as
the same drive letter.
6.
Restart Analysis Services on the query server.
7.
As far as the database is concerned, we have used the Analysis Services 2008 R2
DW database. For better results, the database is partitioned as follows:
StoreInventory measure group
ProductVendor measure group
RetailInventory measure group
8. Now, we are at the point where we need to generate a query workload. To generate
a large number of query streams that are both CPU and I/O intensive, we have used
a sample application tool called ASQueryGenerator from the CodePlex site, which
can be downl oaded from http://www.codeplex.com/SQLSrvAn alysisSrvcs .
9. The scale-out approach is increasing the capacity of OLAP applications, such as
building the OLAP farm.
10. Choose the server with the higher configuration to be a master server. The master
server contains the metadata and manages the whole farm.
11. The processing sever will serve as the front-end server to communicate with clients
and execute client requests.
12. The front-end servers (Analysis query servers) are configured to run under a network
load-balancing cluster. A front-end server needs a fast CPU for making calculations
and a lot of memory to cache the results.
 
Search WWH ::




Custom Search