Data Management performance guidelines white paper > Data Source Components (DSCs)
 
Data Source Components (DSCs)
Underlying the performance of data management operations is the performance of the DSCs that are used for reading, writing, and updating data. This section compares the performance of the most frequently used DSCs and details each DSC’s strengths and weaknesses. The following DSCs are included in the analysis:
Relational Database CDSC
Data File CDSC
Quantum DSC
Quanvert DSC
SPSS SAV DSC
Note The Triple-S DSC is not included in the analysis because it uses the Quantum DSC for reading and writing fixed-format data.
Unless stated otherwise, the performance analysis was carried out using the UNICOM Intelligence Developer Documentation Library skidemo example. The skidemo example has the following characteristics:
Characteristic
Number
Fields
34
Variable instances
77
 
Numeric variables
7
 
Text variables
2
 
Categorical variables
68
Categories
144
Records
3142
Responses
293330
DSC guidelines and summary
Feature
RDB
DDF
Quantum
Quanvert
SAV
Storage type
Database
1 file
1 or 2 files
Folder
1 file
Read
Yes
Yes
Yes
Yes
Yes
Write/update
Yes
Yes
Yes
No
Yes
Unbounded loops (levels)
No
Yes
No
Yes
No
WHERE clause support
Yes
Yes
No
No
No
Compressed format
No
Yes
No
Yes
No
Multiple user read
Yes
Yes
Yes
Yes
Yes
Multiple user write
Yes
No
No
No
No
The recommended DSC usage is as follows:
Web-based or multi-user data collection (Web/CATI)
Use the Relational Database DSC (RDB).
Single-user data collection (CAPI/DE)
Use the Data File DSC (DDF).
Use the SPSS SAV DSC (SAV) when IBM SPSS Statistics is used to manage and analyze data.
Data management tasks
Use the Data File DSC (DDF).
Data export/portable data formats
Use the Quantum or Triple-S DSCs for sharing data between different Market Research applications.
Use the SPSS SAV files for easy sharing of data between different SPSS product suites.
Use the Data File DSC for sharing data that is intended to be used by UNICOM Intelligence applications.
Use the Data File DSC for sharing data that is intended to be used by UNICOM Intelligence applications.
Tabulation/analysis
In almost all cases, leave the data in its original format for tabulation, unless further data processing is required, or if native WHERE clause support is required for improved performance.
Otherwise, use the Data File DSC (DDF).
See
Relational Database (RDB) CDSC
UNICOM Intelligence Data File (DDF) CDSC
Quantum CDSC
Quanvert DSC
SPSS SAV DSC
Data Management performance guidelines white paper