Auto-tools/Projects/ActiveData/Redshift: Difference between revisions

Auto-tools/Projects/ActiveData/Redshift (view source)

No change in size , 19 March 2015

m

Confirmed users

1,927

edits

@@ Line 17: / Line 17: @@
 * Manually managed schema - Redshift is a database, and like most databases, has a fundamentally different philosophy about data than a document store does.  ActiveData should be agnostic about the shape of the data it stores; leaving it to the application endpoints to determine that.  Redshift demands these schemas be managed manually, or a program written to extend the schema as new fields are encountered.
 * JSON is just a string –  Redshift does not index the properties found in JSON, but does provide [http://docs.aws.amazon.com/redshift/latest/dg/json-functions.html JSON functions] so they can be used in a query.  We must pull out the interesting JSON properties to their own columns before the full benefit of Redshift is realized.
-* No concept if ID – There is no UNIQUE checking in Redshift, which requires extra programmer diligence when upserting records.  Even NoSQL solutions have some way to distinguish if a document/record is being overwritten, so I consider this an odd omission by the Redshift team.
+* No concept of ID – There is no UNIQUE checking in Redshift, which requires extra programmer diligence when upserting records.  Even NoSQL solutions have some way to distinguish if a document/record is being overwritten, so I consider this an odd omission by the Redshift team.
 ==Configuration==