Comparison of Product Extensions and Data Sources / Datamart

Deciding whether to use Product Extensions (PX) or Data Sources (DS) / Datamarts is sometimes not easy. This table should help you identify the main differences:

Area	Product Extension	Data Source / Datamart

Area	Product Extension	Data Source / Datamart
Needed license for	Price Setting module	Analytics module
Designed for big data (e.g. transaction history)	No	Yes
Designed for frequently changing data	Yes	No. The data needs to be loaded from Data Feeds to Data Source and e.g. daily flushed to the Data Source.
Can values be edited by the user	Yes	No
Creating charts	Via dashboards using calculation logic	Directly in the UI and from Datamarts only
Data loading – automatic	By IntegrationManager using HTTP transactions (also in batched versions): loaddata/PX integrate/PX delete/PX massupdate/PX add/PX By creating integration in PlatformManager From a CF logic (few records only) using `api.addOrUpdate(), api.update(), api.delete()`	By IntegrationManager using HTTP transactions: loaddata/DMDS From Price Setting as a Data Load using Data Source templates built-in functionality By creating integration in PlatformManager
Data loading – manual	Data upload functionality in PlatformManager By uploading an XLS file in the UI Using Excel Client	Data upload functionality in PlatformManager By uploading an XLS file in the UI (Data tab) Using Excel Client
Data truncation	Manually, by the Mass Delete button By IntegrationManager	Manually, by the Truncate button By Truncate Data Load (supports scheduling) By IntegrationManager
Querying from Groovy	`api.find("PX", Filter.equal("name", "SalesPrice"), Filter.equal(...))` Note: `api.getBatchInfo()` should be used in LPGs and price lists to avoid stability issues	`def ctx = api.getDatamartContext() def ds = ctx.getDataSource("SalesDS") -or- def ds = ctx.getDataMart("SalesDM") def query = ctx.newQuery(ds) .select(..) ...` Note: `api.getBatchInfo()` should be used in LPGs and price lists to avoid stability issues
Selecting distinct values	`api.find("PX", 0, 1, null, ["attribute1"], true, filters)`
Aggregation: SUM, AVG, MIN, MAX, COUNT
Group by	Not available
Joins	No	Natively, Datamart is a join of Data Sources In Groovy, using `executeSqlQuery()` see SD feature: Groovy SQL query API and Datamart Join Sample Code
Exists	Using `Filter.custom`	Using SQL query
Sorting	Yes	Yes
Creating database indexes	Only for customers having dedicated instance; cannot have many of them Recommendation for the main PX index is (attribute1, attribute2) or potentially (attribute1, attribute2, attribute3).	Created for the key fields and for each dimension