SQL Server: No New Index. No Schema Changes. 10x Faster Queries – SQL Server Batch Mode on Rowstore Deep Dive!! Part 4!

Hi guys, welcome to Part 4!!

I hope you are enjoying this series. I truly believe it can be very useful for those who use SQL Server professionally as I have been doing for more than 20 years (how time flies!).

Today we will talk about Batch Mode on Rowstore and how to make your Query run 10x Faster Without Adding Any Index

Well...in Part 3 we introduced Columnstore indexes and saw how Batch Mode execution can dramatically improve analytical queries.

If you missed the previous part, you can read it here:

SQL Server Filtered Index vs Indexed View – Complete Performance Comparison

But here’s the twist.

Starting with SQL Server 2019, you don’t always need a Columnstore index to get Batch Mode performance.

SQL Server can activate Batch Mode on Rowstore.

And sometimes… as i told you at the start of this post ....your query becomes 5x or 10x faster without adding a single index. So i suggest you to continue to read...

What Is Batch Mode?

Traditionally, SQL Server processes queries in Row Mode. That means:

  • One row at a time
  • One operator call per row
  • High CPU overhead for large datasets

Batch Mode changes the execution model.

  • Processes rows in batches (typically up to 900 rows per batch)
  • Uses vectorized processing
  • Reduces CPU cycles per row
  • Improves aggregation and join performance dramatically

Instead of calling operators millions of times, SQL Server processes chunks of rows in a single CPU-efficient operation.

Row Mode vs Batch Mode – The Real Difference

Feature Row Mode Batch Mode
Processing Unit 1 row Up to ~900 rows
CPU Efficiency Lower Much Higher
Aggregation Speed Moderate Very Fast
Memory Usage Standard Optimized

Batch Mode is not just faster. It changes how operators internally execute.

When Does SQL Server Activate Batch Mode on Rowstore?

Batch Mode on Rowstore is available starting from:

  • SQL Server 2019
  • Database compatibility level 150

To enable it: 


ALTER DATABASE YourDatabase
SET COMPATIBILITY_LEVEL = 150;

SQL Server may automatically choose Batch Mode if:

  • The query is analytical (GROUP BY, aggregates, joins)
  • Large number of rows involved
  • Estimated cost justifies vectorization

No Columnstore index required.

Demo – Row Mode vs Batch Mode

Let’s test a large aggregation: 


SELECT CustomerID,
       SUM(Amount) AS TotalAmount
FROM dbo.Sales
GROUP BY CustomerID;

Step 1 – Compatibility Level 140 (SQL Server 2017 behavior)


ALTER DATABASE YourDatabase
SET COMPATIBILITY_LEVEL = 140;

Execution Plan:

  • Hash Aggregate
  • Row Mode
  • High CPU consumption

Step 2 – Compatibility Level 150


ALTER DATABASE YourDatabase
SET COMPATIBILITY_LEVEL = 150;

Execution Plan now shows:

  • Hash Aggregate (Batch Mode)
  • Vectorized execution
  • Lower CPU cost

Performance Comparison

Compatibility Level Execution Mode CPU (ms) Elapsed (ms)
140 Row Mode 1200 1500
150 Batch Mode 220 320

Same query. Same indexes. Same data.

Completely different execution engine behavior.

Memory Grant Feedback

SQL Server 2019 also improves memory management using Memory Grant Feedback.

If the first execution:

  • Overestimates memory → wastes resources
  • Underestimates memory → spills to tempdb

SQL Server adjusts the memory grant for subsequent executions.

Combined with Batch Mode, this creates:

  • More stable performance
  • Fewer spills
  • Better concurrency

Adaptive Joins

Another intelligent feature introduced in modern versions is Adaptive Joins.

Instead of choosing Nested Loop or Hash Join at compile time, SQL Server can decide at runtime based on actual row counts.

This reduces the risk of:

  • Bad cardinality estimates
  • Parameter sniffing side effects
  • Plan instability

Batch Mode works extremely well together with Adaptive Joins.

Why This Matters

Until SQL Server 2017, getting Batch Mode required Columnstore.

From SQL Server 2019 onward, you may gain analytical acceleration without changing your schema.

That means:

  • No additional index maintenance
  • No extra storage
  • No schema redesign

Sometimes performance is not about adding objects.

It’s about enabling the right execution model.

Final Thoughts

Filtered Index reduces I/O.

Indexed View removes aggregation cost.

Columnstore changes the storage engine.

Batch Mode on Rowstore changes the execution engine.

And that is a completely different level of optimization.

In performance tuning, the biggest gains often come not from adding more indexes, but from understanding how SQL Server really executes your queries.

See you in Part 5 😉

Comments

Post a Comment

I Post più popolari

Speaking to Sql Server, sniffing the TDS protocol

Image
Hello friends!  If yesterday you read my post and found it definitely boring (for the topic of course and not for how it was written to the author 😁) well today I promise you that the topic will be much more interesting! Let's talk about Sniffing , let's talk about going to see what happens when we "log in" or when we send a Query to SQL Server.  We will do this by taking a look to the little known tabular data stream (TDS) protocol built on top of the TCP/IP protocol. Ready! The TDS protocol What is the Tabular data stream (TDS) protocol?  The TDS is the protocol used by the SQL Server client Net-Library to communicates with SQL Server . TDS is specific to SQL Server and it is a low-level protocol that specifies both commands and data in a specific arrangement.  TDS encapsulate the TCP/IP protocol. Sniffing TDS packets you can read from the network the requests that your application do to the SQL Server and read the answer .   How to Sniff data from the ne...
Read more

SQL Server, find text in a Trigger, Stored Procedures, View and Function. Two ways and what ways is better

Image
Hi guys, Welcome back! Today a light post about a question that has been asked to me several times. The question is: how to  search for text inside a trigger or a stored procedure ? We have two different ways. A first way using a system view called syscomments: SELECT obj.xtype, text , * FROM sysobjects obj JOIN syscomments com on obj.id = com.id WHERE TEXT like '%text_that_i_am_searching_for%' A second way using a system view called sql_module: SELECT o. name AS Object_Name, o. type_desc , m. definition FROM sys.sql_modules m JOIN sys.objects o ON m. object_id = o. object_id WHERE m. definition Like '%text_that_i_am_searching_for%' ;   Another question is? What differences do we have in using one way instead of the other? Which solution is better? You have to consider that the system view sys.syscomment is obsolete and will be removed sooner or later. So don't use it in production enviroiments. The sys.sql_m...
Read more

SQL Server, Avoid that damn Table Spool!

Image
Hi guys, Welcome back for another post! A light one inspired from what i do during my work: many times I optimize queries rewriting them or adding index and i like it very much! Sometimes you need to use a little imagination, but everything you learn ... sooner or later it will come in handy! Enjoy the reading mates! Avoid the Table Spool, when you can A few days ago I came across a query whose execution time was very high ... or so it was for the client complaining about the problem. Before recreating it, let's create the structures and fill the tables with some data Create table PriceList (Id int identity (1,1) primary Key ,Code Varchar (20)) Insert into PriceList (code) select top 1000 right( '0000' + cast ( row_number () over ( order by id) as varchar (4)),4) from artico Insert into PriceList (code) select top 4000 right( '1000' + cast ( row_number () over ( order by id) as varchar (4)),4) from artico Create table OrdersList (Id int identity...
Read more