The "Transformation" product empowers users to create a Transformation object within the OWOX BI project. A Transformation can consist of at least one Operation or more, forming a sequence of SQL queries. These SQL queries are used to prepare and update data stored in Google BigQuery in your active Google Cloud Platform (GCP) project.
When a user executes a Transformation, it triggers the execution of each internal Operation. Each run of an Operation contributes to a Consumption Unit count. There are two types of Consumption Units:
- Operation Run: The Operation is executed on behalf of OWOX's GCP project, and this is a consumption unit by default for each OWOX BI Project. Processing costs are covered by OWOX company.
- Operation Run Lite: The Operation is performed on behalf of the customer's GCP project. Customers cover processing costs through their GCP billing account.
General information about consumption units in Transformation
- All Operation Runs or Operation Run Lite are rounded to the whole hundred if less within a calendar month.
- There is no difference in the calculation of how the Operation was run - automatically or manually.
- There is no difference in what kind of SQL statement was executed (select, update, delete, etc) for calculating Consumption Units.
How does Operation Run count?
Each run of a separate Operation in a Transformation counts as 1 Operation Run and has a limited processing capacity of up to 20 GB. Exceeding this processing limit is treated as a new Operation Run.
How does Operation Run Lite count?
Each run of a separate Operation in a Transformation counts as 1 Operation Run Lite, regardless of the amount of data processed in the result.
How can you switch between them?
If you wish to change the default Consumption Unit for all Transformations in your OWOX BI Project, please contact us via email at email@example.com.
Examples of calculation consumption units
Example of Operation Run in one Transformation
There are two cases of running the equal Transformation 'Sessions' with three Operations inside. They have different amounts of processed data.
The next table presents the calculation logic of the Operation Run Consumption Unit for these two cases.
|Operation name||Case 1||Case 2|
|Processed data, Gb||Operation Run||Processed data, Gb||Operation Run|
|#1. Preclean data||5||1 (<20Gb)||28||2 (20Gb+8Gb)|
|#2. Preparation data||16||1 (<20Gb)||216||11
|#3. Creating sessions||12||1 (<20Gb)||
|Total Operation Runs in one Transformation run||–||3 (1+1+1)||–||17 (2+11+4)|
So, in this example, we can see that calculation is made by Operation, but not by Transformation. The Transformation just summarizes the result of the Operation Run for all successfully executed Operations.
The result of such calculation is presented in the history log of the Transformation for every Operation and the client is able to see and validate it.
Example of Operation Run for a month
OWOX platform collects the result of every Operation Run calculation and provides this summarized result.
This example uses the previous example for the Transformation 'Session' calculation but for a month period.
|Activity||Case 1||Case 2|
|Everyday Transformation run 'Sessions' for the previous day||
1 Transformation run per day
3 Operation Run per day
Every week Transformation 'Sessions' run for the previous week
|1 Transformation run per week
17 Operation Run per week
|Total Operation Run that was run automatically per month||90 (3×30 days)||68 (17×4 weeks)|
|Manual Transformation run that users made for any reason per month||
3 Transformation run
|2 Transformation run
34 Operation Run
|Total for automatic and manual Transformation run per month||99 Operation Run
90 auto + 9 manual
|102 Operation Run
68 auto + 34 manual
|Grand total Operation Run after rounding up to the monthly bill||100||200|
Example of Operation Run Lite calculation in one Transformation
In the table below, you can see an example of Operation Run Lite calculation compared to Operation Run calculation.
|Operation name||Processed data, Gb||Operation Run Lite, count||Operation Run, count|
|#1. Preclean data||0.5||1||1|
|#2. Preparation data||43||1||3|
|#3. Creating sessions||89||1||5|
|Total (in one Transformation run):||132,5||3||9|