Pages

Differences Between Aggregate and Block Storage

Aggregate storage applications differ from block storage applications in both concept and design.Additionally, aggregate storage applications have some limitations that do not apply to block storage applications. The following tables describe the differences between aggregate and block
storage.
1.Inherent Differences Between Aggregate Storage and Block Storage

Inherent Differences Aggregate Storage Block Storage
Storage kernel Architecture that supports rapid aggregation, optimized to support high dimensionality and sparse data Multiple blocks defined by dense and sparse dimensions and their members, optimized for financial applications
Physical storage definition Through the Application Properties window,Tablespaces tab in Administration Services Through the Database Properties window,Storage tab in Administration Services
Database creation Migrate a block storage outline or define after application creation
Note: Do not use the file system to copy a block storage outline into an aggregate storage application. Use the migration wizard in Administration Services to migrate the outline.
Define after application creation
Databases supported per application One Several (one recommended)
Application and database names Names reserved for tablespaces, cannot be used as application or database names:
● default
● log
● metadata
● temp
Follow the Naming Restrictions for Applications
and Databases
Application and database information display Displayed in the Application Properties window and the Database Properties window in Administration Services. (Information not supported by or relevant to aggregate storage applications is not shown. For a description of aggregate storage specific information, see the Essbase Administration Services Online Help for the Application Properties window and Database Properties window.) Displayed in the Application Properties window and the Database Properties window in Administration Services
Configuration settings (essbase.cfg) For a list of the settings that apply to aggregate storage databases, For a list of the settings that do not apply to block storage databases,
2.Outline Differences Between Aggregate Storage and Block Storage

Outline Functionality Aggregate StorageBlock Storage
Multiple hierarchies enabled, dynamic hierarchy, or stored hierarchy designation Relevant Not relevant
Accounts dimensions and members on dynamic hierarchies Support with the following exceptions:
● No two-pass calculation (however, for
information on specifying the calculation
order.
● No association of attribute dimensions with
the dimension tagged as Accounts
● Additional restrictions for shared members.
Full support
Members on stored hierarchies Support with the following exceptions:
● Support for the ~ (no consolidation)
operator (underneath label-only members
only) and the + (addition) operator
● Cannot have formulas
● Restrictions on label only members
● No Dynamic Time Series members
● Stored hierarchy dimensions cannot have
shared members. Stored hierarchies within
a multiple hierarchies dimension can have
shared members.
Full support
Member storage types Support with the following exceptions:
● Dynamic Calc and Store not relevant
● On stored hierarchies, two limitations if a member is label only:
❍ All dimension members at the same
level as the member must be label only
❍ The parents of the member must be
label only.
Note: On dynamic hierarchies, ability to tag any member as label only
Note: On conversion from a block storage database, attribute dimension members are tagged as Dynamic Calc. On standard dimension members Dynamic Calc tags are converted and tagged as stored members,which changes the Members Stored value on the Dimensions tab of the Database Properties window in Administration Services.
Support for all member storage types in all types of dimensions except attribute dimensions
Ragged hierarchies and hierarchies with more than 10 levels Support, with possible performance impact Support
Outline validation ● When database is started
● When outline is saved
● When block storage outline is converted to
aggregate storage outline
● When user requests
● When outline is saved
● When user requests
Outline paging Support No Support
Database restructure There are several levels of restructure; see “Aggregate Storage Database Restructuring" There are levels of restructure; see “Optimizing Database Restructuring”.
3.Calculation Differences Between Aggregate Storage and Block Storage
Calculation Functionality Aggregate Storage Block Storage
Database calculation Aggregation of the database, which can be predefined by defining aggregate views Calculation script or outline consolidation
Formulas Allowed with the following restrictions:
● Must be valid numeric value expressions written in MDX (cannot contain % operator, replace with expression: (value1/value2)*100)
● No support for Essbase calculation functions
● On dynamic hierarchy members, formulas are allowed without further restrictions
Support for Essbase calculation functions
Calculation scriptsNot supportedsupported
Attribute calculations dimensionSupport for SumSupport for Sum, Count, Min, Max, and Average
Calculation order Member formula calculation order can be defined by the user using the solve order member property Defined by the user in the outline consolidation order or in a calculation script
4.Partitioning and Write Back Differences Between Aggregate Storage and Block Storage
Partitioning and Write-Back Functionality Aggregate Storage Block Storage
Partitioning Support with the following restrictions:
● Transparent partitions and Linked partitions are supported
● Replicated partitions are not supported
● Aggregate storage database as the source database only for transparent partitions
● No outline synchronization
Support with no restrictions
User ability to change data (write back) Transparent partition technique used to enable limited write back Full support
5.Data Load Differences Between Aggregate Storage and Block Storage
Data Load Functionality Aggregate Storage Block Storage
Cells loaded through data loads Only level 0 cells whose values do not depend on formulas in the outline are loaded Cells at all levels can be loaded (except Dynamic Calc members)
Update of database values At the end of a data load, if an aggregation exists, the values in the aggregation are recalculated No automatic update of values. To update data values you must execute all necessary calculation scripts
Data load buffers The loading of multiple data sources into aggregate storage databases is managed through temporary data load buffers. Not supported.
Atomic replacement of the contents of a database When loading data into an aggregate storage database, you can replace the contents of the database or the contents of all incremental data slices in the database. Not supported.
Data slices Aggregate storage databases can contain multiple slices of data. Data slices can be merged.Not supported.
Dimension build for shared members Full support for parent-child build method.Duplicate generation (DUPGEN) build method limited to building alternate hierarchies up to generation 2 (DUPGEN2). Support for all build methods
Loading data mapped to dates In a date-time dimension, you can load data into level-0 members using supported dateformat strings instead of member names. Date-time dimension type is not supported.
6.Query Differences Between Aggregate Storage and Block Storage
Query Functionality Aggregate StorageBlock Storage
Report Writer Supported, except for commands related to sparsity and density of data Fully supported
Spreadsheet Toolkit Supported, with limited ability to change data (write back)Fully supported
APISupportedSupported
Export Support with the following restrictions:
● Export of level 0 data only (no upper-levelexport)
● No columnar export
Supported
MDX queriesSupportedSupported
Queries on attribute members that are associated with non-level 0 members Returns values for descendants of the nonlevel 0 member. Returns #MISSING for descendants of the non-level 0 member
Queries on attribute members and shared members A shared member automatically shares the attribute associations of its non-shared member A shared member does not share the attribute associations of its non-shared member
Query loggingNot supportedsupported
Query performance Considerations when querying data from a dimension that has multiple hierarchies Hierarchies not relevant

0 comments:

© 2010 Datawarehousing Support | Home | Disclaimer | Privacy Policy