Creating Analysis Services Cubes - SQL Server Analysis Services 2012 Cube Development - page 99

analysisservices/2011/engine/
300"xmlns:ddl300_300="http://schemas.microsoft.com/
analysisservices/2011/engine/300/300">
<ID>Fact Reseller Sales</ID>
<Name>Fact Reseller Sales</Name>
<Sourcexsi:type="DsvTableBinding">
<DataSourceViewID>Adventure Works
DW2012</DataSourceViewID>
<TableID>dbo_FactResellerSales</TableID>
</Source>
<StorageMode>Molap</StorageMode>
<ProcessingMode>Regular</ProcessingMode>
<ProactiveCaching>
<SilenceInterval>-PT1S</SilenceInterval>
<Latency>-PT1S</Latency>
<SilenceOverrideInterval>-PT1S</SilenceOverrideInterval>
<ForceRebuildInterval>-PT1S</ForceRebuildInterval>
<Sourcexsi:type="ProactiveCachingInheritedBinding"
/>
</ProactiveCaching>
</Partition>
</ObjectDefinition>
</Create>
Notice that XMLA contains references to a partition's parent objects: database, cube,
and measure group, specifies the partition identifier as well as its name, storage
mode, and processing mode. The partition is based on the FactResellerSales
table, meaning that it will include all the records from this table. You could also bind
a partition to a view or a query. Whether you bind each SSAS partition to a table or
view is a matter of preference and performance. The Partition wizard has the neces-
sary intelligence to warn you that a given fact table is already referenced by an ex-
isting partition, and you cannot use it for additional partitions within the same group;
otherwise, you would include duplicate records, and your solution would report incor-
rect results. Binding a partition to a query offers additional flexibility in a sense that
you can change the SSAS partition without having to make any changes to the rela-
tional objects. Using query binding also allows you to quickly create an empty parti-
tion (by adding a condition, which will never be met, such as SELECT column_1,