1. Developing modern applications with Temporal Tables and JSON
Jos de Bruijn, Borko Novakovic
Senior Program Manager

2. 1 2 3 Takeaways Make your life easier, let the database do it.
Tech Ready 15
Takeaways
4/22/2018
Make your life easier, let the database do it.
Temporal tables enable easy time travel over data.
JSON connects SQL to modern apps and services. 1 2 3
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

3. Less code is better Works but error-prone Simple and intuitive
4/22/2018
Works but error-prone
IF EXISTS
(SELECT * FROM sys.objects
WHERE object_id = OBJECT_ID(N'[dbo].[mytable]') AND TYPE IN (N'U') )
DROP TABLE [dbo].[DumpMe]
Simple and intuitive
DROP IF EXISTS [dbo].[mytable]
© 2015 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

4. Complex tasks made simple
4/22/2018
Analyze data history yourself
Let the temporal tables do it 1
SCHEMA MAINTENANCE
Double effort to maintain current and history. 1
SCHEMA MAINTENANCE
Automatic and online. 2
HISTORY TRACKING
User code is required (triggers, SPs, app).
Hard to maintain and achieve good performance. 2
HISTORY TRACKING
Automatic and optimal. 3
DATA ANALYSIS
Complex queries are required. 3
DATA ANALYSIS
Simple and efficient with
FOR SYSTEM_TIME clause. 4
DATA PROTECTION
Immutability of history data cannot be guaranteed. 4
DATA PROTECTION
Out-of-box immutability of history data.
© 2015 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

5. 4/22/2018
Temporal tables
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

6. When to use temporal tables
4/22/2018
Audit and data forensics
Historical data analysis
Slowly changing dimensions
Pinpoint and correct erroneous data
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

7. How to start with temporal tables
Microsoft Ignite 2015
How to start with temporal tables
4/22/2018 3:42 PM
ANSI compliant
no change in programming model
new insights
SELECT * FROM temporal
Querying
FOR SYSTEM_TIME
AS OF
FROM..TO
BETWEEN..AND
CONTAINED IN
Temporal Querying
INSERT / BULK INSERT
UPDATE
DELETE
MERGE
DML
CREATE temporal TABLE PERIOD FOR SYSTEM_TIME…
ALTER regular_table TABLE ADD PERIOD…
DDL
© 2015 Microsoft Corporation. All rights reserved. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

8. Getting started (DDL) PERIOD definition VERSIONING options
CREATE TABLE Department (
DepNum char(10) NOT NULL PRIMARY KEY CLUSTERED,
DepName varchar(50) NOT NULL,
MngrID INT NULL,
SysStart datetime2 GENERATED ALWAYS AS ROW START,
SysEnd datetime2 GENERATED ALWAYS AS ROW END,
PERIOD FOR SYSTEM_TIME (SysStart,SysEnd)
)WITH(SYSTEM_VERSIONING = ON (HISTORY_TABLE = DepartmentHistory) )
ALTER TABLE Department
ADD ValidFrom datetime2 (0) GENERATED ALWAYS AS ROW START HIDDEN,
ADD ValidTo datetime2 (0) GENERATED ALWAYS AS ROW END HIDDEN,
PERIOD FOR SYSTEM_TIME (ValidFrom, ValidTo);
ALTER TABLE Product
SET (SYSTEM_VERSIONING = ON
(HISTORY_TABLE = dbo.Department_History));
PERIOD definition
VERSIONING
options

9. Temporal System-Versioning (DML)
Microsoft Ignite 2015
Temporal System-Versioning (DML)
4/22/2018 3:42 PM
temporal table (current data)
history table
* Old versions
Insert / Bulk Insert / Select
Update */ Delete *
© 2015 Microsoft Corporation. All rights reserved. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

10. * Include historical versions
Microsoft Ignite 2015
Temporal queries
4/22/2018 3:42 PM
temporal table (current data)
history table
* Include historical versions
Regular queries: SELECT …
Temporal queries*: SELECT … AS OF time
© 2015 Microsoft Corporation. All rights reserved. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

11. Temporal queries (details)
AS OF
“get actual row versions”
BETWEEN..AND
CONTAINED IN
Department (current + history)
Department (current)
DepNum
DepName
MngrID
From To
A001
Marketing 6
2008 ∞
A002
Sales 5
2007
DepNum
DepName
MngrID
A001
Marketing 5 6
A002
Sales 2
A003
Consulting 10
A001 ∞
A001
A002
Department (history) ∞
A002
DepNum
DepName
MngrID
From To
A001
Marketing 5
2005
2008
A002
Sales 2
2007
A003
Consulting 6
2006 10
2009
2012
A003
A003
2005
2016
period of validity
current time
SELECT * FROM Department FOR SYSTEM_TIME
CONTAINED IN (' ', ' ')
SELECT * FROM Department FOR SYSTEM_TIME
BETWEEN ' ' AND ' '
SELECT * FROM Department
SELECT * FROM Department FOR SYSTEM_TIME
AS OF ' '

12. Managing Retention of Historical Data
Stretched history
Partitioned history
(sliding window)
Custom delete script
ALTER TABLE dbo.DepartmentHistory SWITCH PARTITION 1 TO DepartmentHistoryStaging;
ALTER PARTITION FUNCTION [fn_Partition_DepartmentHistory_By_SysEndTime]()
MERGE RANGE(N' T23:59:59.999');
/*...*/
Check out MSDN page for more details
ALTER TABLE dbo.DepartmentHistory
SET (REMOTE_DATA_ARCHIVE = ON (MIGRATION_STATE = OUTBOUND));

13. 4/22/2018
JSON
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

14. Scenarios for JSON in SQL
4/22/2018
Interchange data with apps and services.
Exploit NoSQL agility to extend your app.
Analyze relevant parts of semi-structured data.
How to choose the service?
Mostly relational data model + JSON use cases -> SQL Database.
Only JSON docs -> Document DB.
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

15. What is (not) JSON in SQL
Additional functions for data interchange and flexibility
Not a native data type
Store it as NVARCHAR.
What works with NVARCHAR, works with JSON.
Index with computed columns and B-trees
There is no specific JSON index type.

16. Features Number Date Customer Price Quantity SO43659
4/22/2018
Built-in functions
ISJSON
JSON_VALUE
JSON_MODIFY
JSON_QUERY
OPENJSON
Transforms JSON text to table
[ { "Number":"SO43659", "Date":" T00:00:00" "AccountNumber":"AW29825", "Price":59.99, "Quantity": }, { "Number":"SO43661", "Date":" T00:00:00“ "AccountNumber":"AW73565“, "Price":24.99, "Quantity": } ]
Number
Date
Customer
Price
Quantity
SO43659
T00:00:00
MSFT
59.99 1
SO43661
T00:00:00
Nokia
24.99 3
FOR JSON
Formats result set as JSON text.
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

17. JSON to table @json: Query with OPENJSON function: Input table data:
4/22/2018
@json:
[ { "Order": { "Number":"SO43659", "Date":" T00:00:00" }, "Account": "Microsoft", "Item": { "Price":59.99, "Quantity": } }, { "Order":{ "Number":"SO43661", "Date":" T00:00:00“ }, "Account": “Nokia“, "Item":{ "Price":24.99, "Quantity": } } ]
Query with OPENJSON function:
SELECT *
FROM OPENJSON
WITH (
Number varchar(200) N'$.Order.Number', Date datetime N'$.Order.Date', Customer varchar(200) N'$.Account', Quantity int N'$.Item.Quantity' )
Input table data:
Number
Date
Customer
Quantity
SO43659
T00:00:00
Microsoft 1
SO43661
T00:00:00
Nokia 3
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

18. Table to JSON JSON output: Input table data:
4/22/2018
Input table data:
JSON output:
[ { "Order": { "Number":"SO43659", "Date":" T00:00:00" }, "Account": "Microsoft", "Item": { "Price":59.99, "Quantity": } }, { "Order":{ "Number":"SO43661", "Date":" T00:00:00“ }, "Account": “Nokia“, "Item":{ "Price":24.99, "Quantity": } } ]
Number
Date
Customer
Price
Quantity
SO43659
T00:00:00
MSFT
59.99 1
SO43661
T00:00:00
Nokia
24.99 3
Query with FOR JSON clause:
SELECT Number AS [Order.Number], Date AS [Order.Date],
Customer AS Account,
Price AS 'Item.UnitPrice', Quantity AS 'Item.Qty'
FROM SalesOrder
FOR JSON PATH
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

19. Built-ins and indexing
4/22/2018
CREATE TABLE SalesOrderRecord ( Id int PRIMARY KEY IDENTITY, OrderNumber NVARCHAR(25) NOT NULL, OrderDate DATETIME NOT NULL, SalesOrderItems NVARCHAR(MAX) CONSTRAINT SalesOrderDetails_IS_JSON CHECK ( ISJSON(SalesOrderItems)>0 ), Price AS
JSON_VALUE(SalesOrderItems, '$.Order.Price') )
CREATE INDEX idx_JsonPrice ON SalesOrderRecord(Price)
INCLUDE (Id, OrderNumber, OrderDate)
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

20. Demo Product catalogs 4/22/2018
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

21. More T-SQL enhancements
4/22/2018
More T-SQL enhancements
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

22. STRING_SPLIT FORMATMESSAGE COMPRESS/DECOMPRESS DROP IF EXISTS
4/22/2018
STRING_SPLIT
Splits a string on a provided separator.
FORMATMESSAGE
Constructs a message from sys.messages or provided string.
COMPRESS/DECOMPRESS
Compresses the input expression using GZIP algorithm.
DROP IF EXISTS
Conditional drop for SQL objects.
AT TIME ZONE
Converts input date into a target time zone.
DATEDIFF_BIG
Returns signed big integer.
BULK IMPORT with UTF8
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

23. Timelines and roadmap Temporal tables JSON T-SQL enhancements
4/22/2018
Temporal tables
SQL Server 2016 RTM
SQL Database - in public preview GA
Automated data retention
JSON
SQL Database – in public preview
Native compilation
Column store integration
T-SQL enhancements
New string functions
STRING_AGG
IMPORT FROM Blob storage
© 2014 Microsoft Corporation. All rights reserved. Microsoft, Windows, and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

24. 1 2 3 Takeaways Make your life easier, let the database do it.
Tech Ready 15
Takeaways
4/22/2018
Make your life easier, let the database do it.
Temporal tables enable easy time travel over data.
JSON connects SQL to modern apps and services. 1 2 3
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

25. jodebrui@microsoft.com Tech Ready 15 4/22/2018
© 2012 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.