Advertisements

Archive

Posts Tagged ‘Hekaton’

In-memory enhancements and improvements in SQL Server 2016

June 14, 2015 3 comments

In-Memory tables were introduced in SQL Server 2014 and were also known as Hekaton tables. I’ve written previously about In-memory tables for SQL Server 2014 and you can check in my [previous posts] to know more about these type of tables with some Hands-on examples and demos.
 

–> In-memory tables as new concept in SQL Server 2014 had lot of limitations compared to normal tables. But with the new release of SQL Server 2016 some limitations are addressed and other features have been added for In-Memory tables. These improvements will enable scaling to larger databases and higher throughput in order to support bigger workloads. And compared to previous version of SQL Server it will be easier to migrate your applications to and leverage the benefits of In-Memory OLTP with SQL Server 2016.
 

–> I have collated all the major improvements here in the table below:

SQL Server 2016 - In Memory

* Collation Support
1. Non-BIN2 collations in index key columns
2. Non-Latin code pages for (var)char columns
3. Non-BIN2 collations for comparison and sorting in native modules
 

–> You can check more about In-Memory tables for SQL Server 2016 in MSDN BoL [here].
 

Check the above details explained in the video below:


Advertisements

AdventureWorks 2014 Sample Database Released! – for SQL Server 2014 (Hekaton)

August 5, 2014 1 comment

Sample DataBases AdventureWorks 2014 for SQL Server 2014 has been released and is ready for download, [link].

The sample includes various flavors of samples that you can use with SQL Server 2014, and are:
1. OLTP Database
2. DW Database
3. Tabular Model Database
4. MultiDimensional Model Database

So download these now and start practicing and working on 2014!!!

SQLServer2014_AdventureWorks2014_Sample_DBs
 

–> Check the same demo on YouTube:


Memory Optimized Indexes | Hash vs Range Indexes – SQL Server 2014

December 20, 2013 3 comments

In SQL Server 2014 for In-Memory tables there are lot of changes in DDLs compared with normal Disk Based Tables. In-Memory Tables related changes we’ve seen in previous posts, check [here]. Here we will see Memory Optimized Index related changes and few important things to take care before designing your Tables and Indexes.
 

–> Some of the main points to note are:

1. Indexes on In-Memory tables must be created inline with CREATE TABLE DDL script only.

2. These Indexes are not persisted on Disk and reside only in memory, thus they are not logged. As these Indexes are not persistent so they are re-created whenever SQL Server is restarted. Thus In-Memory tables DO NOT support Clustered Indexes.

3. Only two types of Indexes can be created on In-Memory tables, i.e. Non Clustered Hash Index and Non Clustered Index (aka Range Index). So there is no bookmark lookup.

4. These Non Clustered Indexes are inherently Covering, and all columns are automatically INCLUDED in the Index.

5. Total there can be MAX 8 Non Clustered Indexes created on an In-Memory table.
 

–> Here we will see how Query Optimizer uses Hash & Range Indexes to process query and return results:

1. Hash Indexes: are used for Point Lookups or Seeks. Are optimized for index seeks on equality predicates and also support full index scans. Thus these will only perform better when the predicate clause contains only equality predicate (=).

2. Range Indexes: are used for Range Scans and Ordered Scans. Are optimized for index scans on inequality predicates, such as greater than or less than, as well as sort order. Thus these will only preform better when the predicate clause contains only inequality predicates (>, <, =, BETWEEN).
 

–> Let’s check this by some hands-on code. We will create 2 similar In-Memory tables, one with Range Index and another with Hash Index:

-- Create In-Memory Table with simple NonClustered Index (a.k.a Range Index):
CREATE TABLE dbo.MemOptTable_With_NC_Range_Index
(
    ID INT NOT NULL
        PRIMARY KEY NONCLUSTERED,
    VarString VARCHAR(200) NOT NULL, 
    DateAdded DATETIME NOT NULL
) WITH (
	MEMORY_OPTIMIZED = ON, 
	DURABILITY = SCHEMA_AND_DATA
)
GO

-- Create In-Memory Table with NonClustered Hash Index:
CREATE TABLE dbo.MemOptTable_With_NC_Hash_Index
(
    ID INT NOT NULL
        PRIMARY KEY NONCLUSTERED HASH WITH (BUCKET_COUNT = 10000),
    VarString VARCHAR(200) NOT NULL, 
    DateAdded DATETIME NOT NULL
) WITH (
	MEMORY_OPTIMIZED = ON, 
	DURABILITY = SCHEMA_AND_DATA
)
GO

 

–> Now we will Insert about 10k records on both the tables, so that we have good numbers of records to compare:

DECLARE @i INT = 1

WHILE @i <= 10000
BEGIN
	INSERT INTO dbo.MemOptTable_With_NC_Range_Index 
		VALUES(@i, REPLICATE('a', 200), GETDATE())

	INSERT INTO dbo.MemOptTable_With_NC_Hash_Index 
		VALUES(@i, REPLICATE('a', 200), GETDATE())

	SET @i = @i+1
END

 

–> Now check the Execution Plan by using equality Operator (=) on both the tables:

SELECT * FROM MemOptTable_With_NC_Hash_Index  WHERE ID = 5000 -- 4%
SELECT * FROM MemOptTable_With_NC_Range_Index WHERE ID = 5000 -- 96%

You will see in the Execution Plan image below that Equality Operator with Hash Index Costs you only 4%, but Range Index Costs you 96%.

SQLServer2014_Hash_vs_Range_1
 

–> Now check the Execution Plan by using inequality Operator (BETWEEN) on both the tables:

SELECT * FROM MemOptTable_With_NC_Hash_Index  WHERE ID BETWEEN 5000 AND 6000 -- 99%
SELECT * FROM MemOptTable_With_NC_Range_Index WHERE ID BETWEEN 5000 AND 6000 -- 1%

You will see in the Execution Plan image below that Inequality Operator with Range Index Costs you only 1%, but Hash Index Costs you 99%.

SQLServer2014_Hash_vs_Range_2
 

So, while designing In-Memory Tables and Memory Optimized Indexes you will need to see in future that how you will be going to query that table. It also depends upon various scenarios and conditions, so always keep note of these things in advance while designing your In-Memory Tables.
 

Update: Know more about In-Memory tables:


 


XTP (eXtreme Transaction Processing) with Hekaton Tables & Native Compiled Stored Procedures – SQL Server 2014

December 19, 2013 1 comment

In my previous posts [this & this] I talked about creating Memory Optimized Database, how to create In-Memory Tables & Native Compiled Stored Procedures and what happens when they are created.

Here in this post we will see how FAST actually In-Memory tables & Native Compiled Stored Procedures are, when compared with normal Disk based Tables & Simple Stored Procedures.

I’ll be using the same [ManTest] database used in my previous posts, you can refer to the DDL script [here].
 

–> We will create:

1. One Disk based Table & one simple Stored Procedure which will use this Disk based Table.

2. One In-Memory Table & one Native Compiled Stored Procedure which will use this In-Memory Table.
 

1. Let’s first create a Disk based Table and a normal Stored Procedure:

USE [ManTest]
GO

-- Create a Disk table (non-Memory Optimized):
CREATE TABLE dbo.DiskTable
(
    ID INT NOT NULL
        PRIMARY KEY,
    VarString VARCHAR(200) NOT NULL, 
    DateAdded DATETIME NOT NULL
)
GO

-- Create normal Stored Procedure to load data into above Table:
CREATE PROCEDURE dbo.spLoadDiskTable @maxRows INT, @VarString VARCHAR(200)
AS
BEGIN
	SET NOCOUNT ON

	DECLARE @i INT = 1

	WHILE @i <= @maxRows
	BEGIN
		INSERT INTO dbo.DiskTable VALUES(@i, @VarString, GETDATE())
		SET @i = @i+1
	END
END
GO

 

2. Now create an In-Memory table & a Native Compiled Stored Procedure to load data:

-- Create an In-Memory table:
CREATE TABLE dbo.MemOptTable
(
    ID INT NOT NULL
        PRIMARY KEY NONCLUSTERED HASH WITH (BUCKET_COUNT = 10000),
    VarString VARCHAR(200) NOT NULL, 
    DateAdded DATETIME NOT NULL
) WITH (
	MEMORY_OPTIMIZED = ON, 
	DURABILITY = SCHEMA_AND_DATA
)
GO

-- Create Native Compiled Stored Procedure to load data into above Table:
CREATE PROCEDURE dbo.spLoadMemOptTable @maxRows INT, @VarString VARCHAR(200)
WITH 
	NATIVE_COMPILATION, 
	SCHEMABINDING, 
	EXECUTE AS OWNER
AS
BEGIN ATOMIC
WITH (TRANSACTION ISOLATION LEVEL=SNAPSHOT, LANGUAGE='us_english')

	DECLARE @i INT = 1

	WHILE @i <= @maxRows
	BEGIN
		INSERT INTO dbo.MemOptTable VALUES(@i, @VarString, GETDATE())
		SET @i = @i+1
	END
END
GO

 

–> Now we will try to Load 10k record in above 2 table in various ways, as follows:

1. Load Disk based Table by T-SQL script using a WHILE loop.

2. Load the same Disk based Table by Stored Procedure which internally uses a WHILE loop.

3. Load In-Memory Table by T-SQL script using a WHILE loop.

4. Load the same In-Memory Table by Native Compiled Stored Procedure which internally uses a WHILE loop.
 

–> Working with Disk based Tables:

SET NOCOUNT ON

DECLARE 
	@StartTime DATETIME2,
	@TotalTime INT

DECLARE 
	@i INT,
	@maxRows INT,
	@VarString VARCHAR(200)

SET @maxRows = 10000
SET @VarString = REPLICATE('a',200)

SET @StartTime = SYSDATETIME()
SET @i = 1

-- 1. Load Disk Table (without SP):
WHILE @i <= @maxRows
BEGIN
	INSERT INTO dbo.DiskTable VALUES(@i, @VarString, GETDATE())
	SET @i = @i+1
END

SET @TotalTime = DATEDIFF(ms,@StartTime,SYSDATETIME())

SELECT 'Disk Table Load: ' + CAST(@TotalTime AS VARCHAR) + ' ms (without SP)'

-- 2. Load Disk Table (with simple SP):
DELETE FROM dbo.DiskTable

SET @StartTime = SYSDATETIME()

EXEC spLoadDiskTable @maxRows, @VarString

SET @TotalTime = DATEDIFF(ms,@StartTime,SYSDATETIME())

SELECT 'Disk Table Load: ' + CAST(@TotalTime AS VARCHAR) + ' ms (with simple SP)'

 

–> Working with In-Memory Tables:

-- 3. Load Memory Optimized Table (without SP):
SET @StartTime = SYSDATETIME()
SET @i = 1

WHILE @i <= @maxRows
BEGIN
	INSERT INTO dbo.MemOptTable VALUES(@i, @VarString, GETDATE())
	SET @i = @i+1
END

SET @TotalTime = DATEDIFF(ms,@StartTime,SYSDATETIME())

SELECT 'Memory Table Load: ' + CAST(@TotalTime AS VARCHAR) + ' ms (without SP)'

-- 4. Load Memory Optimized Table (with Native Compiled SP):
DELETE FROM dbo.MemOptTable

SET @StartTime = SYSDATETIME()

EXEC spLoadMemOptTable @maxRows, @VarString

SET @TotalTime = DATEDIFF(ms,@StartTime,SYSDATETIME())

SELECT 'Disk Table Load: ' + CAST(@TotalTime AS VARCHAR) + ' ms (with Native Compiled SP)'
GO

–> Output (Loaded 10k records):

Disk based Table Load	 : 28382 ms (without SP)
Disk based Table SP Load : 8297 ms  (with simple SP)
In-Memory Table Load     : 5176 ms  (without SP)
In-Memory Table SP Load  : 174 ms   (with Native Compiled SP)

 

So, you can clearly see the benefit and multifold increase in performance by using In-Memory Tables & Native Compiled Stored Procedures. The graph below shows performance in visual bar charts, impressive, isn’t it?

SQLServer2014Comparison
 

–> Final Cleanup

DROP PROCEDURE dbo.spLoadDiskTable
DROP TABLE dbo.DiskTable
DROP PROCEDURE dbo.spLoadMemOptTable
DROP TABLE dbo.MemOptTable
GO

 

Update: Know more about In-Memory tables:


 


CTP-2 released for SQL Server 2014 | and I’ve installed it !

October 16, 2013 1 comment

Much awaited Community Test Preview 2 (CTP-2) for SQL Server 2014 is released and you can Download it from [here].

Check out the Release Notes [here]. This lists some limitations, issues and workarounds for them.
 

–> As mentioned in my [previous post] for CTP-1:

– You cannot upgrade your existing installation of CTP-1 to CTP-2.

– and similar to CTP-1 restrictions you cannot install CTP-2 with pre-existing versions of SQL Server, SSDT and Visual Studio.

Sp, this should also be a clean install to be used only for learning and POCs, and should not be used on Production Environments. Installation is very simple and similar to CTP-1 and latest SQL Server previous versions.
 

–> What’s new with CTP-2:

1. Can create Range Indexes for Ordered Scans (along with Hash Indexes in CTP-1).

2. Configure the In-memory usage limit to provide performance and stability for the traditional disk-based workloads.

3. Memory Optimization Advisor wizard added to SSMS for converting disk-based Tables to In-memory (Hekaton) Tables, by identifying Incompatible Data Types, Identity Columns, Constraints, Partitioning, Replications, etc.

4. Similar to above a “Naive Compilation Advisor” wizard for converting Stored Procedures to Natively Compiled SPs, by identifying Incompatible SQL statements, like: SET Options, UDFs, CTE, UNION, DISTINCT, One-part names, IN Clause, Subquery, TVFs, GOTO, ERROR_NUMBER, INSERT EXEC, OBJECT_ID, CASE, SELECT INTO, @@rowcount, QUOTENAME, EXECUTE, PRINT, EXISTS, MERGE, etc.

5. and many more enhancements with Always On like: allowing to view XEvents in UTC time, triggering XEvents when replicas change synchronization state, and recording the last time and transaction LSN committed when a replica goes to resolving state, new wizard to greatly simplify adding a replica on Azure.
 

Enough for now, let me go back and work with CTP-2, wait for more updates !!!