Large dimension tables

I am trying to get a grip with how a dimension table can have more records than the fact table since the key for the dimension table is DIMID. Can someone give me a practical example? Thanks

Thanks for this lets work with the Doc No example.
1) I have a cube with 4 dimensions - the fourth dimension contains material no, doc no and plant, the key for this dimension table is DIMID (system generated ??) and each record in the Dimension contains the corresponding SID values for mat no, doc no and plant
2) My first load of data into the cube - 1000 records
3) For each of these 1000 records a DIMID is generated in dimension 4 and the corresponding values of the 3 SIDs are retained as part of these records
4) My next load is a delta 200 lines - 150 new records and 50 updated records - in this case 150 new DIMIDs are generated and the 3 SIDS are retained whereas the original 50 updated records do not have new DIMIDs generated
Is how I described the scenario correct

Similar Messages

Non Unique Indexes on Dimension Table

Hello ,
I have Material Dimension that has Product MainGroup, Brand , SPC code broken from Product Hierarchy.
As per Business Requirement , we don't want to use the Product Hierarchy , that should need to split into 3 pieces.
Since The Cube Dimensions already reached the 13 , we can not increase the dimensions to keep these 3 fields into separate dimensions.
I heard from Basis guy as <b>we can have index on three fields in same dimension table</b> & read some negative impact on aggregate definition.
Is it true , which one is true , am not sure.
Which one impact causes more worst & usefull..?
Could please some experts throw some light on this.
Cheers
Martin

Martin -
Not sure what you mean by "read some negative impact on aggregate definition".
But as far as adding additional indices on other columns of a dimension table, that certainly is doable. I have never done this as part of an actual intentiional design, but it seems like a valid apporach if you are limited by dimensions. I'm assuming when you say you already have 13, that the 13 does not include the three standard dimensions for time, request, (drawing a blank, is it currency), so that you really have 16 dimensions, 13 of which are user defined.
We have added dimension tabl e indices in our shop when we have found large dimension tables that, either from poor initial design, or changes to the data and/or queries, have resulted in full tables scans against large dimension tables. In some cases, the query costs of the full scan of the dimension table was more than the cost for the access of the fact table itself.
These indices must be added by your DBA as they can not be added thru the Admin Wkbench. You should also probably keep a record of any of these indices you create because if for some reason you delete the tables and reactivate the cube, you'll probably lose them.
Pizzaman

Dimension table is larger than the fact table

Hi Community,
How can we explain the phenomenon when a dimension table has MORE records in it than the fact table ? What are the conditions that would cause this to occur ?
Thank you !
Keith

Thanks, Bhanu,
I am wondering specifically how to explain the output from program SAP_INFOCUBE_DESIGNS when the dimension table is shown to have a fact table ratio that is greater than 100%
I believe that SAP_INFOCUBE_DESIGNS already takes into consideration both the E and also the F-fact table when calculating the ratio. So in this case, we could not explain it by your first suggestion (after compression - but looking at only the F table).
In the case where selective deletions have been performed, how can we correct the situation ? For example, how could we clean out the records in the dimension tables which no longer have any facts in the fact table ? (I think the BW system should do this automatically as a part of the selective deletion, don't you agree ?).
Also, is there any other explanation for how the dimension table could arrive at greater than 100% the size of the fact table(s) ?
For example, lets say that (theoretically) we placed many very dynamic characteristics together into the same dimension.. which we know you should not do. Would it be possible for the combination of these very many dynamic characteristics to cause so many DIM IDs that the dimension table overtakes the record count of the fact table ? Is this situation then made worse by compression if the number of fact table records is reduced thanks to removal of the request ID ?

Multiple columns from the same dimension table as row labels performing slowly

(Working with SSAS tabular)
I'm trying to figure out what the approach should be for the following scenario:
Lets say we have a Customer table. The table has columns such as account number, department number, name, salesperson, account manager, number of customers, delivery route, etc
A user of the model could want to see any permutation of that information as the row labels. How should that be handled?
What we've been doing so far is that the user adds each column they want into the "ROWS" section in Excel. This works fine with smaller tables (for example, "Department" table with a "Department Code" and "Department Name",
but on large tables this quickly chokes. I understand why this is happening, I just haven't found a better way to accomplish the same thing.
I can add a calculated column to the model through VS, but obviously this is unsupportable and unscalable when each person needs their own permutations of the data. Can something similar be done in Excel?
This question seems to be what I need:
http://social.msdn.microsoft.com/Forums/en-US/97d1157a-1402-4227-b96a-79524401ddcd/mdx-query-performance-when-selecting-multiple-attributes-from-same-dimension?forum=sqlanalysisservices
However I can't find any information on how to add those properties (is it a multidimensional-only thing?)

Thanks for the help. Sorry but i'm a self-taught developer, and i may be missing some basics :)
Anyway i've done what you suggested but i get this error:
[nQSError: 15011]The dimension table source Dimension Services.DM_D_SERVIZI_SRV has an aggregate content specification that specifies the level Product. But the source mapping contains column COD_PRODUCT with a functional dependency association on a more detailed level .
where:
- DM_D_SERVIZI_SRV is the physical alias for the Service Dimension (and the name of the LTS too)
- COD_PRODUCT is the leaf of the hierarchy, the physical primary key, but it hasnt to be included in the hierarchy
Do i have to add another level with the primary key and hide it to the users?
I tried to solve this going to the logical tables source properties, on the tab contents, setting "logical level" to null for the hierarchy, but i don't know if this is correct.
Thanks

Multiple Fact Tables and Dimension Tables

I have been having some problems trying to model the data from Oracle E-Business Suite maintenance. I will try to give the best description of how the data is held in the tables. The structure is such that a work order can have multiple operations and an operation can have multiple resources as well. I believe the problem comes in the fact that an operation doesn't necessarily need to have a resource. I could not attach an image so I have written out an example below. I am not saying this is right or that it works, but just to give you an idea of what I am thinking. The full dimension would be Organization -> WorkOrder -> Operation -> Resource. Now, the fact tables all hold factual data for the three different levels, with the facts being at each corresponding level. This causes an obvious problem in combining the tables into one large fact table through the ETL process.
Can anyone tell me if they think this can be done? Am I way off? I am sure that there is a solution as there always is but I have been killing myself trying to figure this one out. I currently have the entire solution in different Business Models. I would like however to be able to compare facts from multiple areas such as the Work Order level and the Resource level.
Any help is greatly appreciated. I realize that the solution may also require additional work on the ETL side so I am open to any and all suggestions.
Thank you in advance for anyones time. :)
Dimension Tables
WorkOrder_D
Operation_D
Resource_D
Organization_D
Fact Tables
WorkOrder_F
Operation_F
Resource_F
Joins
WorkOrder_D -> Operation_D
Operation_D -> Resource_D
WorkOrder_D -> WorkOrder_F
Operation_D -> Operation_F
Resource_D -> Resource_F
Organization_D -> WorkOrder_D
Organization_D -> Operation_D
Organization_D -> Resource_D

Hi,
Currently the dimension table is taken as a simple logical table in rpd as it does not have have any levels or hierarchy.
Its a flat dimension. Can you guide me how can I implement a flat dimension in OBIEE? Because this dimension is taken as simple logical table
I am not able to set appropriate level for fac tables. This dimension does not appear in the list of dimensions.

Dimension Table populating data

Hi
I am in the process of creating a data mart with a star schema.
The star schema has been defined with the fact and dimension tables and the primary and foreign keys.
I have written the script for one of the dimensions and would like to know when the job runs on a daily basis should the job truncate the table every day and rebuild the dimension table or should it only add new records to the table. If it should add only
new records to the table how do is this done?
I assume that the fact table job is run once a day and only new data is added to it?
Thanks

It will depend on the volume of your dimensions. In most of our projects, we do not truncate, we update only updated rows based on a fingerprint (to make the comparison faster than column by column), and insert for new rows (SCD1). For SCD2 we apply
similar approach for updates and inserts, and expirations in batch (one UPDATE for all applicable rows at the end of the package/ETL).
If your dimension is very large, you can consider truncating all data or deleting only affected modified rows (based on nosiness key) to later reload those, but you have to be carefully maintaining the same surrogate keys reference by your
existing facts.
HTH,
Please, mark this post as Answer if this helps you to solve your question/problem.
Alan Koo | "Microsoft Business Intelligence and more..."
http://www.alankoo.com

Dimension key 16 missing in dimension table /BIC/DZPP_CP1P

Hi all,
I have a problem with an infocube ZPP_CP1. I am not able to delete nor load any data. It was working fine till some time back.
Below is the outcome of running RSRV check on this cube. I tried to run the error correction in RSRV. But n o use
Please help.
Dimension key 16 missing in dimension table /BIC/DZPP_CP1P
Message no. RSRV018
Diagnosis
The dimension key 16 that appears as field KEY_ZPP_CP1P in the fact table, does not appear as a value of the DIMID field in the dimensions table /BIC/DZPP_CP1P.
There are 17580 fact records that use the dimension key 16.
The facts belonging to dimension key 16 are therefore no longer connected to the master data of the characteristic in dimension.
Note that errors that are reported for the package dimension are not serious (They are thus shown as warnings (yellow) and not errors (red). When deleting transaction data requests, it can arise that the associated entries in the package dimension have already been deleted. As a result, the system terminates when deleting what can be a very large number of fact records. At the moment, we are working on a correction which will delete such data which remains after deletion of the request. Under no circumstances must you do this manually. Also note that data for request 0 cannot generally be deleted.
The test investigates whether all the facts are zero. If this is the case, the system is able to remove the inconsistency by deleting these fact records. If the error cannot be removed, the only way to re-establish a consistent status is to reconstruct the InfoCube. It may be possible for SAP to correct the inconsistency, for which you should create an error message.
Procedure
This inconsistency can occur if you use methods other than those found in BW to delete data from the SAP BW tables (for example, maintaining tables manually, using your own coding or database tools).

Hi Ansel,
There has been no changes in the cube. I am getting this problem in my QA server. So I retransported the cube again from Dev to QA. But did not help me..
Any other ideas??
Regards,
Adarsh

DWH: how do you analyze fact and dimension tables

Hi,
I was wondering how you analyze your fact and dimension tables.
Our fact table is partitioned per month. Each partition contains 4M rows and is 270 MB large. We are using 9 dimensions, 6 have about 50'000 rows (2MB), 1 about 1M rows (50MB) and 2 about 3M rows (200MB). All tables are compressed. The version of Oracle we are using is 9.2.0.5.
What I was wondering is how you would, using dbms_stats, analyze the fact and dimension tables. Which percentage would you analyze? On which column would you build histograms?

nope, but i could copy-paste the URL or I could copy-paste the entire thread from the other forum. Id did the one that made more sense to me.

Dimension table and fact table exists data physically

Hi experts,
can anyone plz tell me weather dimension table and fact table exists data physically or not/

Hi..Sudheer
SAPu2019s BW is based on "Enhanced Star schema" or "Info Cubes" database design.This database design has a central database table, known as u2018Fact Tableu2019 which is surrounded by associated dimension tables.
Fact table is surrounded by dimensional tables. Fact table is usually very large, that means it contains
millions to billions of records.
These dimension tables doesn't contain data it contain references to the pointer tables that point to the master data tables which in turn contain Master data objects such as customer, material and destination country stored in BW as Info objects. An InfoObjects can contain single field definitions such as transaction data or complex Customer Master Data that hold attributes, hierarchy and customer texts that are stored in their own tables.
SID is surrogate ID generated by the system. The SID tables are created when we create a master data IO. In SAP BW star schema, the distinction is made between two self contained areas: Infocube & master data tables/SID tables.
The master data doesn't reside in the satr schema but resides in separate tables which are shared across all the star schemas in SAP BW. A numer ID is generated which connects the dimension tables of the infocube to that of the master data tables.
The dimension tables contain the dim ID and SID of a particular IO. Using this SID the attributes and texts of an master data Io is accessed.
The SID table is connected to the associated master data tables via teh char key.
Fact table(Transaction data,DIM ID)<>Dimention Table(SID and Dim ID)<->Masterdata table(SID,IO)
Thanks,
Abha

Fact and dimension table partition

My team is implementing new data-warehouse. I would like to know that when should we plan to do partition of fact and dimension table, before data comes in or after?

Hi,
It is recommended to partition Fact table (Where we will have huge data). Automate the partition so that each day it will create a new partition to hold latest data (Split the previous partition into 2). Best practice is to create partition on transaction
timestamps so load the incremental data into a empty table called (Table_IN) and then Switch that data into main table (Table). Make sure your tables (Table and Table_IN) should be on one file group.
Refer below content for detailed info
Designing and Administrating Partitions in SQL Server 2012
A popular method of better managing large and active tables and indexes is the use of partitioning. Partitioning is a feature for segregating I/O workload within
SQL Server database so that I/O can be better balanced against available I/O subsystems while providing better user response time, lower I/O latency, and faster backups and recovery. By partitioning tables and indexes across multiple filegroups, data retrieval
and management is much quicker because only subsets of the data are used, meanwhile ensuring that the integrity of the database as a whole remains intact.
Tip
Partitioning is typically used for administrative or certain I/O performance scenarios. However, partitioning can also speed up some queries by enabling
lock escalation to a single partition, rather than to an entire table. You must allow lock escalation to move up to the partition level by setting it with either the Lock Escalation option of Database Options page in SSMS or by using the LOCK_ESCALATION option
of the ALTER TABLE statement.
After a table or index is partitioned, data is stored horizontally across multiple filegroups, so groups of data are mapped to individual partitions. Typical
scenarios for partitioning include large tables that become very difficult to manage, tables that are suffering performance degradation because of excessive I/O or blocking locks, table-centric maintenance processes that exceed the available time for maintenance,
and moving historical data from the active portion of a table to a partition with less activity.
Partitioning tables and indexes warrants a bit of planning before putting them into production. The usual approach to partitioning a table or index follows these
steps:
1. Create
the filegroup(s) and file(s) used to hold the partitions defined by the partitioning scheme.
2. Create
a partition function to map the rows of the table or index to specific partitions based on the values in a specified column. A very common partitioning function is based on the creation date of the record.
3. Create
a partitioning scheme to map the partitions of the partitioned table to the specified filegroup(s) and, thereby, to specific locations on the Windows file system.
4. Create
the table or index (or ALTER an existing table or index) by specifying the partition scheme as the storage location for the partitioned object.
Although Transact-SQL commands are available to perform every step described earlier, the Create Partition Wizard makes the entire process quick and easy through
an intuitive point-and-click interface. The next section provides an overview of using the Create Partition Wizard in SQL Server 2012, and an example later in this section shows the Transact-SQL commands.
Leveraging the Create Partition Wizard to Create Table and Index Partitions
The Create Partition Wizard can be used to divide data in large tables across multiple filegroups to increase performance and can be invoked by right-clicking
any table or index, selecting Storage, and then selecting Create Partition. The first step is to identify which columns to partition by reviewing all the columns available in the Available Partitioning Columns section located on the Select a Partitioning Column
dialog box, as displayed in Figure 3.13. This screen also includes additional options such as the following:
Figure 3.13. Selecting a partitioning column.
The next screen is called Select a Partition Function. This page is used for specifying the partition function where the data will be partitioned. The options
include using an existing partition or creating a new partition. The subsequent page is called New Partition Scheme. Here a DBA will conduct a mapping of the rows selected of tables being partitioned to a desired filegroup. Either a new partition scheme should
be used or a new one needs to be created. The final screen is used for doing the actual mapping. On the Map Partitions page, specify the partitions to be used for each partition and then enter a range for the values of the partitions. The
ranges and settings on the grid include the following:
Note
By opening the Set Boundary Values dialog box, a DBA can set boundary values based on dates (for example, partition everything in a column after a specific
date). The data types are based on dates.
Designing table and index partitions is a DBA task that typically requires a joint effort with the database development team. The DBA must have a strong understanding
of the database, tables, and columns to make the correct choices for partitioning. For more information on partitioning, review Books Online.
Enhancements to Partitioning in SQL Server 2012
SQL Server 2012 now supports as many as 15,000 partitions. When using more than 1,000 partitions, Microsoft recommends that the instance of SQL Server have at
least 16Gb of available memory. This recommendation particularly applies to partitioned indexes, especially those that are not aligned with the base table or with the clustered index of the table. Other Data Manipulation Language statements (DML) and Data
Definition Language statements (DDL) may also run short of memory when processing on a large number of partitions.
Certain DBCC commands may take longer to execute when processing a large number of partitions. On the other hand, a few DBCC commands can be scoped to the partition
level and, if so, can be used to perform their function on a subset of data in the partitioned table.
Queries may also benefit from a new query engine enhancement called partition elimination. SQL Server uses partition enhancement automatically if it is available.
Here’s how it works. Assume a table has four partitions, with all the data for customers whose names begin with R, S, or T in the third partition. If a query’s WHERE clause
filters on customer name looking for ‘System%’, the query engine knows that it needs only to partition three to answer
the request. Thus, it might greatly reduce I/O for that query. On the other hand, some queries might take longer if there are more than 1,000 partitions and the query is not able to perform partition elimination.
Finally, SQL Server 2012 introduces some changes and improvements to the algorithms used to calculate partitioned index statistics. Primarily, SQL Server 2012
samples rows in a partitioned index when it is created or rebuilt, rather than scanning all available rows. This may sometimes result in somewhat different query behavior compared to the same queries running on SQL Server 2012.
Administrating Data Using Partition Switching
Partitioning is useful to access and manage a subset of data while losing none of the integrity of the entire data set. There is one limitation, though. When
a partition is created on an existing table, new data is added to a specific partition or to the default partition if none is specified. That means the default partition might grow unwieldy if it is left unmanaged. (This concept is similar to how a clustered
index needs to be rebuilt from time to time to reestablish its fill factor setting.)
Switching partitions is a fast operation because no physical movement of data takes place. Instead, only the metadata pointers to the physical data are altered.
You can alter partitions using SQL Server Management Studio or with the ALTER TABLE...SWITCH
Transact-SQL statement. Both options enable you to ensure partitions are
well maintained. For example, you can transfer subsets of data between partitions, move tables between partitions, or combine partitions together. Because the ALTER TABLE...SWITCH statement
does not actually move the data, a few prerequisites must be in place:
• Partitions must use the same column when switching between two partitions.
• The source and target table must exist prior to the switch and must be on the same filegroup, along with their corresponding indexes,
index partitions, and indexed view partitions.
• The target partition must exist prior to the switch, and it must be empty, whether adding a table to an existing partitioned table
or moving a partition from one table to another. The same holds true when moving a partitioned table to a nonpartitioned table structure.
• The source and target tables must have the same columns in identical order with the same names, data types, and data type attributes
(length, precision, scale, and nullability). Computed columns must have identical syntax, as well as primary key constraints. The tables must also have the same settings for ANSI_NULLS and QUOTED_IDENTIFIER properties.
Clustered and nonclustered indexes must be identical. ROWGUID properties
and XML schemas must match. Finally, settings for in-row data storage must also be the same.
• The source and target tables must have matching nullability on the partitioning column. Although both NULL and NOT
NULL are supported, NOT
NULL is strongly recommended.
Likewise, the ALTER TABLE...SWITCH statement
will not work under certain circumstances:
• Full-text indexes, XML indexes, and old-fashioned SQL Server rules are not allowed (though CHECK constraints
are allowed).
• Tables in a merge replication scheme are not allowed. Tables in a transactional replication scheme are allowed with special caveats.
Triggers are allowed on tables but must not fire during the switch.
• Indexes on the source and target table must reside on the same partition as the tables themselves.
• Indexed views make partition switching difficult and have a lot of extra rules about how and when they can be switched. Refer to
the SQL Server Books Online if you want to perform partition switching on tables containing indexed views.
• Referential integrity can impact the use of partition switching. First, foreign keys on other tables cannot reference the source
table. If the source table holds the primary key, it cannot have a primary or foreign key relationship with the target table. If the target table holds the foreign key, it cannot have a primary or foreign key relationship with the source table.
In summary, simple tables can easily accommodate partition switching. The more complexity a source or target table exhibits, the more likely that careful planning
and extra work will be required to even make partition switching possible, let alone efficient.
Here’s an example where we create a partitioned table using a previously created partition scheme, called Date_Range_PartScheme1.
We then create a new, nonpartitioned table identical to the partitioned table residing on the same filegroup. We finish up switching the data from the partitioned table into the nonpartitioned table:
CREATE TABLE TransactionHistory_Partn1 (Xn_Hst_ID int, Xn_Type char(10)) ON Date_Range_PartScheme1 (Xn_Hst_ID) ; GO CREATE TABLE TransactionHistory_No_Partn (Xn_Hst_ID int, Xn_Type
char(10)) ON main_filegroup ; GO ALTER TABLE TransactionHistory_Partn1 SWITCH partition1 TO TransactionHistory_No_Partn; GO
The next section shows how to use a more sophisticated, but very popular, approach to partition switching called a sliding
window partition.
Example and Best Practices for Managing Sliding Window Partitions
Assume that our AdventureWorks business is booming. The sales staff, and by extension the AdventureWorks2012 database, is very busy. We noticed over time that
the TransactionHistory table is very active as sales transactions are first entered and are still very active over their first month in the database. But the older the transactions are, the less activity they see. Consequently, we’d like to automatically group
transactions into four partitions per year, basically containing one quarter of the year’s data each, in a rolling partitioning. Any transaction older than one year will be purged or archived.
The answer to a scenario like the preceding one is called a sliding window partition because
we are constantly loading new data in and sliding old data over, eventually to be purged or archived. Before you begin, you must choose either a LEFT partition function window or a RIGHT partition function window:
1. How
data is handled varies according to the choice of LEFT or RIGHT partition function window:
• With a LEFT strategy, partition1 holds the oldest data (Q4 data), partition2 holds data that is 6- to 9-months old (Q3), partition3
holds data that is 3- to 6-months old (Q2), and partition4 holds recent data less than 3-months old.
• With a RIGHT strategy, partition4 holds the holds data (Q4), partition3 holds Q3 data, partition2 holds Q2 data, and partition1
holds recent data.
• Following the best practice, make sure there are empty partitions on both the leading edge (partition0) and trailing edge (partition5)
of the partition.
• RIGHT range functions usually make more sense to most people because it is natural for most people to to start ranges at their lowest
value and work upward from there.
2. Assuming
that a RIGHT partition function windows is used, we first use the SPLIT subclause of the ALTER PARTITION FUNCTIONstatement
to split empty partition5 into two empty partitions, 5 and 6.
3. We
use the SWITCH subclause
of ALTER TABLE to
switch out partition4 to a staging table for archiving or simply to drop and purge the data. Partition4 is now empty.
4. We
can then use MERGE to
combine the empty partitions 4 and 5, so that we’re back to the same number of partitions as when we started. This way, partition3 becomes the new partition4, partition2 becomes the new partition3, and partition1 becomes the new partition2.
5. We
can use SWITCH to
push the new quarter’s data into the spot of partition1.
Tip
Use the $PARTITION system
function to determine where a partition function places values within a range of partitions.
Some best practices to consider for using a slide window partition include the following:
• Load newest data into a heap, and then add indexes after the load is finished. Delete oldest data or, when working with very large
data sets, drop the partition with the oldest data.
• Keep an empty staging partition at the leftmost and rightmost ends of the partition range to ensure that the partitions split when
loading in new data, and merge, after unloading old data, do not cause data movement.
• Do not split or merge a partition already populated with data because this can cause severe locking and explosive log growth.
• Create the load staging table in the same filegroup as the partition you are loading.
• Create the unload staging table in the same filegroup as the partition you are deleting.
• Don’t load a partition until its range boundary is met. For example, don’t create and load a partition meant to hold data that is
one to two months older before the current data has aged one month. Instead, continue to allow the latest partition to accumulate data until the data is ready for a new, full partition.
• Unload one partition at a time.
• The ALTER TABLE...SWITCH statement
issues a schema lock on the entire table. Keep this in mind if regular transactional activity is still going on while a table is being partitioned.
Thanks Shiven:) If Answer is Helpful, Please Vote

Fact and dimension table relationship?

hi
in se38 i executed this program sap_infocube_designs
i got all cubes and percentage , this is directly fact and dimension table relationship based on this i need to take action is it line item dimension or high cardinality (dimen>20% fact line, dimen>10<20 fact is high cardinality.
regards
suneel.

hi,
line item has to be choosen in such a way to control the dim table size for the char that have almost large unique records.
Line item dim table will not be shown by this program.
Ramesh

Oracle Data Compression on SID tables and Dimension Tables

Hello Community,
We have had great success with Oracle compression on ODS tables that are no longer loaded.
We'd now like to move on to other types of BW tables that are very large.
OSS Note 701235 provides answers to questions concerning the possible use of Oracle compression together with SAP BW.
But the Note does not give suggestions for (or against) Oracle compression on SID tables or Dimension tables.
I believe both table types would exhibit the same behaviour : mostly inserts of new SID IDs and new DIM IDs, but few updates to existing SID or Dimension records. If this is true, then both are good candidates for oracle compression.
Do you also agree that this is the typical behaviour for SID tables and dimension tables ? And that these types of tables are good candidates for Oracle compression in a large BW system ?
Thanks kindly!
Keith Helfrich

Hi all,
Although this is an old thread I stumbled on during my own investigations I can provide some answers to your questions.
Table candidates for compression are found by these criteria
       - Table size big enough?
       - Long lifetime of object planned ?
      - No or only rare structural changes for the table   ?
      - u201EUpdateu201C rate low : is your data mostly kind of u201Cread onlyu201D ?
--               for the wideley used rolling window partition techniques of
                  tables in BW this is not a problem: mostly INSERTu2019s in the
                 current partition not affecting other partitions
BW tables that can benefit from compression (see SAP notes 105047,701235)
       - PSA tables with data that must be saved for a longer time
       - ODS change log (no Updates of old data, only Inserts of new data)
       - u201Ehistoricalu201C cubes wich get no changes in table structure anymore
Limitations
         - normal Insert or Update statements are stored ALWAYS in uncompressed
                format and must be compressed separately ( <= Oracle10g )
         - Slight CPU overhead of compression, butu2026
         --      CPU consumption is more than compensated by doing less I/O as
               for Bulk loads or parallel processing
         --      SAP BW transformations took a significant amount of CPU for
                   overall load-time into cubes caused by the application server not
                   the database
          - Table must have not more than 255 fields
          - Adding columns with an initial value or dropping columns require
               uncompression of the complete table (strongest limitation)
Consider all this above and you can decide that tables that going through UPDATE's are
not good candidates for compression or tables that can change it's structure (like
Fact- or DIM tables) .
Now, my questions to you:
Wich Oracle version do you use?
Wich tool do you use for Oracle compression?
BRspace (can you give an example ?)
ALTER ... MOVE COMPRESS
bye
yk

Logical sql firing larger aggregate table instead of smaller one

Hi
When we process a request containing one particular column alone along with some time dimension say month or year, the logical sql is firing larger aggregare table instead of smaller aggregate table. Please, help us in resolving this issue.
The OracleBI version we are using is 10.1.3.4.1
Thanks.

Hi,
Try posting in the OLAP forum.
OLAP
Thanks, Mark

#datasync error - could that come from merging on fields from fact tables instead of dimensions tables?

I'm reporting out of two universes published by Epic
1. Warehouse - Patient
2. Warehouse - Transactions
I wanted to join two queries based on the primary key for a hospital encounter. Following the tutorial seemed pretty straightforward until I got to displaying data from a merged query.
The table displayed results from Query1, but adding fields from Query2 wiped out all the data in the table, leaving only #datasync in each field.
My workaround to get fields from both queries displayed (see screenshot)
1. Merge queries on *two* fields - primary key of hospital encounter and primary key of patient
2. Create new variable
3. Make variable type Detail
4. Associate variable with hospital encounter key from Query1
5. Set formula equal to a field from Query2
I'm not sure why this workaround works or if what I'm experiencing is a symptom of something larger. Could this workaround be needed because I am merging on fields from a fact table instead of fields from a dimension table?
Thanks in advance

You have defined merged dimension on 'company code-Key' which is common in both the Bex-queries.
You are able to bring the merged dimension and other objects from Query_ One to report block without any issue but when adding objects from Query_Two you are getting the error #DATASYNC.
In this case objects from Query_One are got sync with the merged dimension object without any issue because they got added first to it.
Similarly when you add merged dimension and objects from Query_Two you find no issue, because objects from Query_Two go sync first.
Once objects from a query(Query_One/Query_Two) got added to a merged dimension, when we try adding objects from other query we get #DATASYNC error. This is because data in the other query is not able to sync with the initial result set, this is a know behavior.
There are two workarounds:
1) Merging all common dimension/characteristic objects: Only merged dimensions data will sync with the initial query, un-merged dimensions/characteristic objects will still give #DATASYNC error.
2) Create detail objects/attribute objects at report level for all uncommon characteristic/dimension objects from query_two referring to merged dimension. Then add these newly created detail/attribute objects to the table block that is having initial query objects with merged dimension, with this you see result set of query_two in the table block not #DATASYNC error.
~ Manoj

Large editable tables

I have a datatable with input fields assigned to each column. The user can edit values freely and then submit the changes for update. Because of the size of the list it appears that a huge amount of processing is required even though only a couple of rows are actually modified by the user.
Is there a "best practices" for handling large editable tables that anyone is aware of?
Thanks

Presuming Oracle backend database.....(you didnt say)
Partitioning would seem a quick (and expensive $$$$) win, if you partition on date, be sure to use the same date key to perform 'partition pruning' , which is to say give the optimizer a chance to throw away all unrequired partitions before it starts going to disk. When you say 'didnt improve performance much' you need to give a partition wise join for the partitioning to become effective at query time - did you do this ? for example a dashboard prompt with a filter on week / month would help (assuming your week / month time dim column is joined on partition key)
Is your data model a star schema ? Do you have dimension metadata in your database ? I'd be looking at Materialized views and looking for some sort of aggregate rollup query re-write whereby you queries are re-written at run time to the Mview and does not hit your fact table.
As for caching, OBIEE caching is the last resort after you exhaust every other alternative, its not there to speed up queries and you would be hiding the problem, what when a user drills down or needs to run an ad-hoc report that is not cached ?
I would start with understanding Oracle execution plans, review your data model (with the view on speedy data extraction) and look at what Oracle DB gives for data warehousing appliances :
parallelism is your friend!
star transformation,
Mviews (table or Cube based)
OLAP

Large dimension tables

Similar Messages

Maybe you are looking for