Teradata 15 New Features
Brief overview of some of the new features added in Teradata 15 version.
➠ Three new string functions were added
Visit String functions
to check syntax and examples for above functions
➠ Teradata JSON:
With TD 15, Teradata started supporting the storage and processing of JSON
- A JSON datatype is either stored as a JSON document or as JSON records in relational format.
- Most of the available functions and procedures can be used for processing, shredding, and publishing JSON data.
- Maximun size of JSON documents can be 16MB.
- JSONPath provides simple traversal and regular expressions with wildcards to filter and navigate complex JSON documents.
- Teradata allows users to extract values from JSON documents and store those values in normal datatypes using JSON shredding functionality
- Teradata allows users to publish results of SQL queries in JSON format using JSON shredding functionality.
- Teradata allows users to collect Statistics on extracted portions of the JSON datatype.
- Existing Teradata join indexing structure is supported for extracted portions of the JSON
data type to provide superior performance.
- Only UNICODE and LATIN character sets are available for the JSON datatype.
- JSON cannot be used with FastLoad/MultiLoad/FastExport due to restrictions on all LOB types. But if the size of JSON is less than 64K then it can be type cast to a
predefined types (VARCHAR / CHAR) and utilities can be used.
- A JSON type cannot be used in a GROUP BY, ORDER BY, PARTITION BY, WHERE, ON, SET, DISTINCT, HAVING, QUALIFY, IN, CUBE, GROUPING SETS or ROLLUP clause.
- There is no implicit mechanism to convert XML to JSON.
➠ 1 MB Phase 2:
The maximum size of several internal Teradata structures has been increased to 1MB.
- Users can specify up to 1MB of parameters in the IN list of a WHERE clause.
- Users can have IN parameter lists of up to 1MB in size for row triggers, stored procedures and other related objects(like table functions).
- Users may observe improved performance and/or disk space usage, when data blocks larger than 128KB will be modified.
➠ 1 GB Sort Implementation:
Large amounts of memory (up to 1 GB per AMP) for SORT can be set in RECONFIG.
Source Teradata Site:
- More rows can be processed as it increases the size of memory used by the in-memory append buffers.
Check Teradata 15 Release Summary
for more detail.