From Glitchdata
Jump to navigation Jump to search

The following table highlights the philosophies in approaching OTLP and OLAP data systems.

# Characteristic OTLP OLAP
1 Source of Data Operational data. OLTPs are typically the original source of data. Consolidated data. OLAP data comes from various OLTP databases.
2 Purpose of Data To manage business operations. To help with planning, business analytics and decision support
3 Data Insights Reveals state of ongoing business processes Multi-dimensional views of various kinds of business activities
4 Inserts and Updates Short and fast inserts and updates initiated by the end user. Executed as periodic long-running batch jobs. Modern business analytics are much faster as the lines between OTLP and OLAP are blurred.
5 Queries Relatively standadized and simple queries. Returns relatively few records. Often complex queries involving aggregations

6 Processing Speed Typically very fast. Depending on the amount of data and datasets. Complex queries can take hours to complete. Processing speed can be improved in various ways. One of which is the creation of indexes.
7 Space Requirements Can be relatively small if historical data is archived. Large due to aggregation structures, and historical data. Also requires more indexes than OLTP.
8 Database Design Highly normalized. Numerous tables. Typically de-normalized with fewer tables. Use of star and snowflake schemas.
9 Backup and Recovery Backups are crucial. Operational data is critical to run a business. Data loss is likely to have significant impact. Instead of regular backups, some environments may consider reloading OLTP data as the recovery mechanism