sp_executesql – Page 63

Fixing Ordered Column Store Sorting In SQL Server 2022

Posted on July 12, 2022July 13, 2022 by Erik Darling

Groove Is In The Heart

When Brent posted about the availability of, and disappointment with creating ordered column store indexes in SQL Server 2022, I got to work.

I can’t have my dear friend Brent being all distraught with all those fast cars around. That’s how accidents happen, and I fear he might leave the Blitz scripts to me in his will or something.

Anyway, I decided to dig in and see what was going on behind the scenes. Which of course, means query plans, and bothering people who are really good at debuggers.

Most of the problems that you’ll run into in SQL Server will come from sorting data.

Whenever I have to think about Sorts, I head to this post about all the different Sorts you might see in a query plan.

Cod Piece

In Paul’s post, he talks about using undocumented trace flag 8666 to get additional details about Sort operators.

Let’s do that. Paul is smart, though he is always completely wrong about which season it is.

DROP TABLE IF EXISTS
    dbo.Votes_CCI;

SELECT
    v.*
INTO dbo.Votes_CCI
FROM dbo.Votes AS v;

I’m using the Votes table because it’s nice and narrow and I don’t have to tinker with any string columns.

Strings in databases were a mistake, after all.

DBCC TRACEON(8666);
CREATE CLUSTERED COLUMNSTORE INDEX
    vcci
ON dbo.Votes_CCI
ORDER (Postid);
DBCC TRACEOFF(8666);

Here’s what we get back in the query plan:

We’ve got a Soft Sort! What does our seasonally maladjusted friend say about those?

A “soft sort” uses only its primary memory grant and never spills. It doesn’t guarantee fully-sorted output. Each sort run using the available memory grant will be sorted. A “sort sort” represents a best effort given the resource available. This property can be used to infer that a Sort is implemented with CQScanPartitionSortNew without attaching a debugger. The meaning of the InMemory property flag shown above will be covered in part 2. It does not indicate whether a regular sort was performed in memory or not.

Well, with that attitude, it’s not surprising that there are so many overlapping buckets in the column store index. If it’s not good enough, what can you do?

Building the index with the Soft Sort here also leads to things being as bad as they were in Brent’s post.

Insert Debugging Here

Alas, there’s (almost) always a way. Microsoft keeps making these trace flag things.

There are a bunch of different ways to track them down, but figuring out the behavior of random trace flags that you may find just by enabling them isn’t easy.

One way to tie a trace flag to a behavior is to use WinDbg to step through different behaviors in action, and see if SQL Server checks to see if a trace flag is enabled when that behavior is performed.

If you catch that, you can be reasonably sure that the trace flag will have some impact on the behavior. Not all trace flags can be enabled at runtime. Some need to be enabled as startup options.

Sometimes it’s hours and hours of work to track this stuff down, and other times Paul White (b|t) already has notes on helpful ones.

The trace flag below, 2417, is present going back to SQL Server 2014, and can help with the Soft Sort issues we’re seeing when building ordered clustered column store indexes today.

Here’s another one:

DBCC TRACEON(8666, 2417);
CREATE CLUSTERED COLUMNSTORE INDEX
    vcci
ON dbo.Votes_CCI
ORDER (Postid)
WITH(MAXDOP = 1);
DBCC TRACEOFF(8666, 2417);

The MAXDOP 1 hint isn’t strictly necessary. With a parallel plan, you may see up to DOP overlapping row groups.

That’s why it was a popular maneuver to emulate this behavior by creating a clustered row store index, and then create a clustered column store index over it with drop existing and a MAXDOP 1 hint.

At DOP 1, you don’t see that overlap. It takes a lot longer of course — 3 minutes instead of 30 or so seconds — which is a real bummer. But without it, you could see DOP over lapping rowgroups.

If you want All The Pretty Little Rowgroups, this is what you have to do.

Anyway, the result using sp_BlitzIndex looks a lot better now:

EXEC sp_BlitzIndex
    @TableName = 'Votes_CCI';

SQL Server Query Results — capture the flag

How nice.

You can also use undocumented and unsupported trace flag 11621, which is

[A] feature flag for the ‘partition sort on column store order’ so the end result is similar, but via a different mechanism to 2417.

A partition sort is useful in general to prevent unnecessary switching between partitions. If you sort the stream by partition, you process all the rows for one before moving on to the next. A soft sort is ok there because it’s just a performance optimization. Worst case, you end up switching between partitions quite often because the sort ran out of memory, but correct results will still occur.

Chain Gang

A “reasonable” alternative to trace flags maybe to adjust the index create memory configuration option. If we set it down to the minimum value, we get a “helpful” error message:

EXEC sys.sp_configure 
    'index create memory', 
    704;

RECONFIGURE;

As promised:

Msg 8606, Level 17, State 1, Line 31

This index operation requires 123208 KB of memory per DOP.

The total requirement of 985800 KB for DOP of 8 is greater than the sp_configure value of 704 KB set for the advanced server configuration option “index create memory (KB)”.

Increase this setting or reduce DOP and rerun the query.

If you get the actual execution plan for the clustered column store index create or rebuild with the Soft Sort disabled and look at the memory grant, you get a reasonable estimate for what to set index create memory to.

Changing it does two things:

Avoids the very low memory grant that Soft Sorts receive, and causes the uneven row groups
The Soft Sort keeps the index create from going above that index create memory number

Setting index create memory for this particular index creation/rebuild to 5,561,824 gets you the nice, even row groups (at MAXDOP 1) that we saw when disabling the Soft Sort entirely.

Bottom line, here is that uneven row groups happen with column store indexes when there’s a:

Parallel create/rebuild
Low memory grant create/rebuild

If this sort of thing is particularly important to you, you could adjust index create memory to a value that allows the Soft Sort adequate memory.

But that’s a hell of a lot of work, and I hope Microsoft just fixes this in a later build.

Reality Bites

The cute thing here is that, while this syntactical functionality has been available in Azure Cloud Nonsense© for some time, no one uses that, so no one cares.

The bits for this were technically available in SQL Server 2019 as well, but I’m not telling you how to do that. It’s not supported, and bad things might happen if you use it.

I mean, bad things happen in SQL Server 2022 where it’s supported unless you use an undocumented trace flag, but… Uh. I dunno.

This trace flag seems to set things back to how things worked in the Before Times, though, which is probably how they should have stayed.

Thanks for reading!

Going Further

If this is the kind of SQL Server stuff you love learning about, you’ll love my training. I’m offering a 75% discount to my blog readers if you click from here. I’m also available for consulting if you just don’t have time for that and need to solve performance problems quickly.

Considerations For Paging Queries In SQL Server With Batch Mode (Don’t Use OFFSET/FETCH)

Posted on July 11, 2022June 25, 2022 by Erik Darling

First Things First

The first SQL Server blog posts that I ever read while trying to solve a specific problem here these two:

They sort of changed my life a little, despite the author’s aversion to the letter Z. So that’s cool. Can’t have everything.

To this day, though, I see people screw up paging queries in numerous ways.

Selecting all the columns in one go
Adding in joins when exists will do
Sticking a DISTINCT on there just because
Thinking a view will solve some problem
Piles and piles of UDFs
Local variables for TOP or OFFSET/FETCH
Not paying attention to indexing

It’s sort of like every other query I see, except with additional complications.

Especially cute for a query slathered in NOLOCK hints is the oft-accompanying concern that “data might change and people might see something weird when they query for the next page”.

Okay, pal. Now you’re concerned.

Modern Love

A while back I recorded a video about using nonclustered column store indexes to improve the performance of paging queries:

While a lot of the details in there are still true, I want to talk about something slightly different today. While nonclustered column store indexes make great data sources for queries with unpredictable search predicates, they’re not strictly necessary to get batch mode anymore.

With SQL Server 2019, you can get batch mode on row store indexes, as long as you’re on Enterprise Edition, and in compatibility level 150.

Deal with it.

The thing is, how you structure your paging queries can definitely hurt your chances of getting that optimization.

Saddened Face

The bummer here is that the paging technique that I learned from Paul’s articles (linked above) doesn’t seem to qualify for batch mode on row store without a column store index in place, so they don’t make the demo cut here.

The good news is that if you’re going to approach this with any degree of hope for performance, you’re gonna be using a column store index anyway.

The two methods we’re going to look at are OFFSET/FETCH and a more traditional ROW_NUMBER query.

As you may have picked up from the title, one will turn out better, and it’s not the OFFSET/FETCH variety. Especially as you get larger, or go deeper into results, it becomes a real boat anchor.

Anyway, let’s examine, as they say in France.

Barfset Wretch

This is the best way of writing this query that I can come up with.

DECLARE 
    @page_number int = 1,
    @page_size int = 1000;

WITH 
    paging AS
(
    SELECT 
        p.Id
    FROM dbo.Posts AS p
    ORDER BY 
        p.LastActivityDate, 
        p.Id 
    OFFSET ((@page_number - 1) * @page_size) 
    ROW FETCH NEXT (@page_size) ROWS ONLY
)
SELECT 
    p.*
FROM paging AS pg
JOIN dbo.Posts AS p
    ON pg.id = p.Id
ORDER BY 
    p.LastActivityDate,
    p.Id
OPTION (RECOMPILE);

Note that the local variables don’t come into play so much here because of the recompile hint.

Still, just to grab 1000 rows, this query takes just about 4 seconds.

This is not so good.

Examine!

Hero Number

The better-performing query here with the batch mode on row store enhancement(!) is using a single filtered ROW_NUMBER to grab the rows we care about.

DECLARE 
    @page_number int = 1,
    @page_size int = 1000;

WITH 
    fetching AS
(
    SELECT 
        p.Id, 
        n = 
            ROW_NUMBER() OVER 
            ( 
                ORDER BY
                    p.LastActivityDate, 
                    p.Id 
            )
    FROM dbo.Posts AS p
)
SELECT 
    p.*
FROM fetching AS f
JOIN dbo.Posts AS p
    ON f.Id = p.Id
WHERE f.n > ((@page_number - 1) * @page_size)
AND   f.n < ((@page_number * @page_size) + 1)
ORDER BY 
    p.LastActivityDate,
    p.Id
OPTION (RECOMPILE);

Again, this is about the best I can write the query. Maybe you have a better way. Maybe you don’t.

Mine takes a shade under 2 seconds. Twice as fast. Examine!

I’ll take twice as fast any day of the week.

Compare/Contrast

The OFFSET/FETCH query plan is all in row mode, while the ROW_NUMBER query has batch mode elements.

You can see this by eyeballing the plan: it has a window aggregate operator, and an adaptive join. There are other batch mode operators here, but none have visual cues in the graphical elements of the plan.

This is part of what makes things faster, of course. The differences can be even more profound when you add in the “real life” stuff that paging queries usually require. Filtering, joining, other sorting elements, etc.

Anyway, the point here is that how you write your paging queries from the start can make a big difference in how they end up, performance-wise.

Newer versions of SQL Server where certain behaviors are locked behind heuristics (absent column store indexes being present in some manner) can be especially fickle.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: Does Not Care To Fix Your Local Variable Problems

Posted on July 8, 2022June 24, 2022 by Erik Darling

–To Fix Parameter Sniffing

There are some code comments you see that really set the stage for how tuning a query is going to go.

Usually one misgiving about how SQL Server works gives way to a whole levee-breaking bevy of other ones and three days later you can’t feel your legs but dammit it’s done.

Okay, maybe it was three hours, but it felt like three days. Something about the gravitation pull of these black hole queries.

One fix I’ve been wishing for, or wish I’ve been fixing for, is a cure for local variables. I’d even be cool if Forced Parameterization was that cure, but you know…

Time will tell.

Husk

Let’s say we’ve got this stored procedure, which does something similar to the “I’m gonna fix parameter sniffing with a local variable hey why is everything around me turning to brimstone before my very eyes?” idea, but with… less of an end-of-times vibe.

CREATE OR ALTER PROCEDURE 
    dbo.IndexTuningMaster
( 
    @OwnerUserId int,
    @ParentId int, 
    @PostTypeId int 
)
AS
BEGIN
SET NOCOUNT, XACT_ABORT ON;

    /*Someone passed in bad data and we got a bad query plan,
      and we have to make sure that doesn't happen again*/
    
    DECLARE 
        @ParentIdFix int = 
            CASE 
                WHEN @ParentId < 0 
                THEN 0 
                ELSE @ParentId 
            END;
    
    SELECT TOP (1) 
        p.*
    FROM dbo.Posts AS p
    WHERE p.ParentId = @ParentIdFix
    AND   p.PostTypeId = @PostTypeId
    AND   p.OwnerUserId = @OwnerUserId
    ORDER BY 
        p.Score DESC, 
        p.Id DESC;

END;

How bad could a top 1 query be, anyway?

Fortune Teller

When we run this query like so and so:

EXEC dbo.IndexTuningMaster 
    @OwnerUserId = 22656, 
    @ParentId = 0, 
    @PostTypeId = 1;

EXEC dbo.IndexTuningMaster 
    @OwnerUserId = 22656, 
    @ParentId = 184618, 
    @PostTypeId = 2;

We come up with zip zero zilch none nada:

We get a super low guess for both. obviously that guess hurts a large set of matched data far worse than a small one, but the important thing here is that both queries receive the same bad guess.

This is a direct side effect of the local variable’s poor estimate, which PSP isn’t quite yet ready to go up against.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: How PSP Can Help Some Queries With IF Branches

Posted on July 7, 2022June 24, 2022 by Erik Darling

Time Served

I’ve spent a bit of time talking about how IF branches can break query performance really badly in SQL Server.

While the Parameter Sensitive Plan (PSP) optimization won’t fix every problem with this lazy coding habit, it can fix some of them in very specific circumstances, assuming:

The parameter is eligible for PSP
The parameter is present across IF branches

We’re going to use a simple one parameter example to illustrate the potential utility here.

After all, if I make these things too complicated, someone might leave a comment question.

The horror

IFTTT

Here’s the procedure we’re using. The point is to execute one branch if @Reputation parameter is equal to one, and another branch if it equals something else.

In the bad old days, both queries would get a plan optimized at compile time, and neither one would get the performance boost that you hoped for.

In the good news days that you’ll probably get to experience around 2025, things are different!

CREATE OR ALTER PROCEDURE 
    dbo.IFTTT 
(
    @Reputation int
)
AS 
BEGIN
SET NOCOUNT, XACT_ABORT ON;

SET STATISTICS XML ON;  

    IF @Reputation = 1
    BEGIN
        SELECT 
            u.Id, 
            u.DisplayName, 
            u.Reputation, 
            u.CreationDate
        FROM dbo.Users AS u
        WHERE u.Reputation = @Reputation;
    END;

    IF @Reputation > 1
    BEGIN
        SELECT 
            u.Id, 
            u.DisplayName, 
            u.Reputation, 
            u.CreationDate
        FROM dbo.Users AS u
        WHERE u.Reputation = @Reputation;
    END;

SET STATISTICS XML OFF; 

END;
GO

Johnson & Johnson

If we execute these queries back to back, each one gets a new plan:

EXEC dbo.IFTTT 
    @Reputation = 1;
GO 

EXEC dbo.IFTTT 
    @Reputation = 2;
GO

Optimize For You

The reason why is in the resulting queries, as usual. The Reputation column has enough skew present to trigger the PSP optimization, so executions with differently-bucketed parameter values end up with different plans.

option (PLAN PER VALUE(QueryVariantID = 3, predicate_range([StackOverflow2013].[dbo].[Users].[Reputation] = @Reputation, 100.0, 1000000.0)))

option (PLAN PER VALUE(QueryVariantID = 2, predicate_range([StackOverflow2013].[dbo].[Users].[Reputation] = @Reputation, 100.0, 1000000.0)))

And of course, each plan has different compile and runtime values:

If I were to run this demo in a compatibility level under 160, this would all look totally different.

This is one change I’m sort of interested to see the play-out on.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: Sometimes There’s Nothing To Fix

Posted on July 6, 2022June 24, 2022 by Erik Darling

Best Intentions

After seeing places where the Parameter Sensitive Plan (PSP) optimization quite stubbornly refuses to kick in, it’s somewhat amusing to see it kick in where it can’t possibly have any positive impact.

Even though some parameters are responsible for filtering on columns with highly skewed data, certain other factors may be present that don’t allow for the type of plan quality issues you might run into under normal parameter sensitivity scenarios:

Adequate indexing
Row goals
Other filtering elements

This isn’t to say that they can always prevent problems, but they certainly tend to reduce risks much of the time.

If only everything were always ever perfect, you know?

Setup

Let’s start by examining some data in the Posts table.

First, PostTypeIds:

Questions and answers are the main types of Posts. The data is clearly skewed, here, and in my testing this does qualify for PSP on its own.

The thing is, there are several attributes that Questions can have that Answers can’t. One of those is a ParentId. Looking through how the top 15 or so of those counts breaks down:

Okay, so! Wikis don’t have ParentIds, neither do Moderator Nominations. More importantly, Questions don’t.

The Question with the Most answers is Id 184618, with 518. A far cry from the next-nearest Post Types, and light years from the number of Questions with a ParentId of zero.

More important than loving your data is knowing your data.

To Query A Butterfly

Let’s say we have this query:

SELECT TOP (5000)
    p.Id,
    p.OwnerUserId,
    p.Score
FROM dbo.Posts AS p
WHERE p.PostTypeId = @po
AND   p.ParentId = @pa
ORDER BY 
    p.Score DESC;

The three things we care about getting done are:

Filtering to PostTypeId
Filtering to ParentId
Ordering by Score

Either of these indexes would be suitable for that:

CREATE INDEX 
    popa
ON dbo.Posts
(
    PostTypeId,
    ParentId,
    Score DESC
)
WITH
(
    SORT_IN_TEMPDB = ON,
    DATA_COMPRESSION = PAGE
);

CREATE INDEX 
    papo
ON dbo.Posts
(
    ParentId,
    PostTypeId,
    Score DESC
)
WITH
(
    SORT_IN_TEMPDB = ON,
    DATA_COMPRESSION = PAGE
);

With No PSP At All

Under compatibility level 150, we can run the query in a variety of ways and get nearly identical performance results:

There’s a 27 millisecond difference between the two to find the first 5000 rows that match both predicates. You would have to run these in a very long loop to accumulate a meaningful overall difference.

In this case, both queries use and reuse the same execution plan. You can see that in the estimates.

With All The PSP

Switching to compat level 160, the queries are injected with the PLAN PER VALUE hint.

SELECT TOP (5000)
    p.Id,
    p.OwnerUserId,
    p.Score
FROM dbo.Posts AS p
WHERE p.PostTypeId = @po
AND   p.ParentId = @pa
ORDER BY 
    p.Score DESC 
OPTION 
(
    PLAN PER VALUE
    (
        QueryVariantID = 2, 
        predicate_range
        (
            [StackOverflow2013].[dbo].[Posts].[PostTypeId] = @po, 
            100.0, 
            10000000.0
        )
    )
)

SELECT TOP (5000)
    p.Id,
    p.OwnerUserId,
    p.Score
FROM dbo.Posts AS p
WHERE p.PostTypeId = @po
AND   p.ParentId = @pa
ORDER BY 
    p.Score DESC 
OPTION 
(
    PLAN PER VALUE
    (
        QueryVariantID = 3, 
        predicate_range
        (
            [StackOverflow2013].[dbo].[Posts].[PostTypeId] = @po, 
            100.0, 
            10000000.0
        )
    )
)

The thing is, both queries end up with identical execution times to when there was no PSP involved at all.

In other words, there is no parameter sensitivity in this scenario, despite there being skew in the column data.

Even searching for the “big” result — Questions with a ParentId of zero, finishes in <30 milliseconds.

Ah well. Gotta train the models somehow.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: Indexing Matters

Posted on July 5, 2022July 5, 2022 by Erik Darling

Starter Pistol

I started writing this post to answer a different question about if the Parameter Sensitive Plan (PSP) optimization worked with computed columns.

Of course it does! Even non-persisted computed columns have statistics generated on them. The only gatekeeping activity from there is whether or not your statistics qualify for the PSP optimization to kick in.

My excitement for the PSP optimization reminds me of the first time I drank Johnnie Walker Blue. I was all riled up on the hype around a $200 bottle of scotch.

Then it tasted like dessert syrup.

That’s a lot like what seeing this pop up in related Extended Events feels like, even for columns with massive amounts of skew in them.

SkewnessThresholdNotMet — how i didn’t meet your mother

Microsoft really need to introduce a hint that tells the optimizer to treat a parameter as being sensitive, and/or a way to tweak the threshold for Skewness not being met.

The heuristics are really falling short, here.

Hash And Rehash

Anyway, let’s use the example that I had started with here, to illustrate that the PSP optimization does work with a computed column, but… like any other column, indexes make all the difference.

I’m using the same example query over and over again, because a lot of the other great examples of parameter sensitivity that I have demo queries written for don’t seem to trigger it.

Here’s the computed column I started with:

ALTER TABLE 
    dbo.Posts 
ADD 
    WhatIsIt 
AS ISNULL
   (
       ParentId,
       0
   );

I’m not actually sure that there are NULL values in there, but the column is marked as NULLable so the ISNULL function doesn’t short-circuit.

Here’s the stored procedure we’ll be using:

CREATE OR ALTER PROCEDURE 
    dbo.FindMeSomePostsPlease 
(
    @ParentId int
)
AS 
SET NOCOUNT, XACT_ABORT ON;
BEGIN

    SELECT
        c = COUNT_BIG(*)
    FROM dbo.Posts AS p
    WHERE ISNULL(p.ParentId, 0) = @ParentId

END;

Bruise the lard that our optimizer can match expressions on its own.

No Index, No Cry

Using the calls that have worked to trigger the PSP optimization in the past, we do not get two important things:

Different cardinality estimates between executions
The option(plan per value... text added to the end of the plan

EXEC dbo.FindMeSomePostsPlease 
    @ParentId = 0;

EXEC dbo.FindMeSomePostsPlease 
    @ParentId = 184618;

The plans are expectedly disappointing:

Indexable

With the help of our trusty ~~steed~~ index, we get the PSP optimization:

CREATE INDEX 
    p
ON dbo.Posts
(
    WhatIsIt
)
WITH
(
    SORT_IN_TEMPDB = ON,
    DATA_COMPRESSION = PAGE
);

Granted, this is a toy example, but…

Bottom line: If you want cool new features to (sometimes) work, you’re gonna need to pay attention to indexes.

What Else About Indexes?

Well, two things. In the example above, PSP didn’t kick in until we had one, I think owing to the computed column not being materialized, but…

Sometimes you’ll get PSP without an index, but the plans won’t be much different aside from cardinality estimates

Here’s a quick example.

DECLARE
    @sql nvarchar(MAX) = N'',
    @po int = 2,
    @pa int = 184618
--184618
SELECT 
    @sql += N'  
SELECT TOP (5000)
    p.Id,
    p.OwnerUserId,
    p.Score
FROM dbo.Posts AS p
WHERE p.PostTypeId = @po
AND   p.ParentId = @pa
ORDER BY 
    p.Score DESC;
';

EXEC sys.sp_executesql
    @sql,
  N'@po int,
    @pa int',
    @po,
    @pa;

First, with no indexes:

Even though both queries have no choice but to scan the clustered index, we get slightly different plans for vastly different estimates. Sort of (ha ha ha) interestingly, if the Sort hadn’t spilled in the top plan, they would have around the same execution time.

Anyway, I’ve been experimenting with other things, but the results have been surprising. For instance, I thought key column order would push the optimizer towards marking one parameter or the other as sensitive, but that doesn’t happen.

Wild stuff.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: Does PSP Work With Dynamic SQL?

Posted on July 4, 2022July 7, 2022 by Erik Darling

No, Really

When I talk to clients about using dynamic SQL, they’re usually under the misconception that those plans can’t get reused.

That may be true under some circumstances when:

It’s not properly parameterized
You use EXEC only and not sp_executesql

Under more favorable circumstances, dynamic SQL gets run, executed, and plans cached and reused with the same frequency as stored procedures.

Now, dynamic SQL isn’t exactly the same as stored procedures. There’s a lot you can do with those that just looks a mess in dynamic SQL, especially longer bits of code.

In today’s post, we’re going to look at how the Parameter Sensitive Plan (PSP) optimization works with dynamic SQL.

Bright, Sunshiny

I just learned how to spell “sunshiny”. Don’t let anyone ever tell you there’s nothing left to learn.

To keep up the sunshiny visage of today’s post, let’s get a TL;DR here: PSP does work with parameterized dynamic SQL.

Here’s an example, using a query with a parameter eligible for the PSP optimization.

DECLARE
    @sql nvarchar(MAX) = 
        N'',
    @parameters nvarchar(MAX) = 
        N'@ParentId int';

SELECT 
    @sql += N'
SELECT
    c = COUNT_BIG(*)
FROM dbo.Posts AS p
WHERE p.ParentId = @ParentId;
';

EXEC sys.sp_executesql
    @sql,
    @parameters,
    0;

EXEC sys.sp_executesql
    @sql,
    @parameters,
    184618;

Both executions here get the option(plan per value... text at the end that indicates PSP kicked in, along with different query plans as expected.

Being Dense

Writing the not-good kind of dynamic SQL, like so:

SELECT 
    @sql = N'
SELECT
    c = COUNT_BIG(*)
FROM dbo.Posts AS p
WHERE p.ParentId = ' + CONVERT(nvarchar(11), 0) + ';';

You will of course get different execution plans, but you’ll get a new execution plan for every different value that gets passed in. You will not get the PSP optimization.

This is not a good example of how you should be writing dynamic SQL. Please don’t do this, unless you have a good reason for it.

Anyway, this is good news, especially for parameterized ORM queries that currently plague many systems in crisis that I get to see every week.

Fun.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: A Missed Opportunity For PSP To Make A Query Go Faster

Posted on July 1, 2022June 24, 2022 by Erik Darling

Farce

The index and proc creation script for this are a bit on the long side, so I’m going to stick them in a GitHub gist to keep the post concise, since we have some other ground to cover here.

I want to lead with the statistics object that gets used for the query, which has a single equality predicate on the parameter value to search the VoteTypeId column.

The relevant portion of the query is this:

FROM     dbo.Votes AS v
WHERE    v.VoteTypeId = @VoteTypeId

The histogram matches the row counts from the table precisely. Thanks, full scan!

So, what’s the problem?

Autoproblematic

To my eye, there’s sufficient skewness here to present parameter sensitivity issues.

But SQL Server disagrees!

733 records vs. 37 million records seems appropriately skewed to me, but we get all of the typical parameter sensitivity symptoms.

Plansplosion

Let’s get ready to rumble, etc.

EXEC dbo.VoteSniffing 
    @VoteTypeId = 4;

EXEC dbo.VoteSniffing 
    @VoteTypeId = 2;

Here are the plans:

Take a moment here to admire the somewhat confusing plan timing that Batch Mode operations are presenting here: The bottom query runs for around 32 seconds.

Up to the Nested Loops join, we use ~17 seconds of wall clock time with our serial execution plan. The Hash Match operator runs for 15 seconds on its own, in Batch Mode.

Parameter sensitivity still happens in SQL Server 2022.

Reversi

If we run the procedure in reverse order, using 4 to cache the plan and then 2 to reuse, we get a similar regression:

Now, okay, let’s pause for a minute here. I need to backtrack, but I don’t want to spend a ton of blogspace on it. I’m also going to put the information in a very snooty block quote.

If we re-run the procedure from the section up above to search for VoteTypeId 2 a second time, Memory Grant Feedback will fix the spill at the Hash Join, and bring the total execution time down to about 15 seconds.

That is an improvement, but… Look at the plan here. If VoteTypeId 2 uses a plan more suited to the number of rows it has to process, the overall time is around 4 seconds, with no need for a memory grant correction.

The second plan for this execution sequence, searching for VoteTypeId 4 second in order, the overall time goes from 0 seconds and 0 milliseconds to 1.4 seconds. The big plan does not make searching for infrequent values faster.

So you see, the big plan isn’t always better.

Missed Connections

Hey, look, this is the first CTP. Maybe stuff like this is still getting ironed out. Maybe this blog post will change the course of history.

I am only moderately influential in the eyes of Microsoft, though, so perhaps not.

Anyway, this seems like a situation with sufficient skew to produce the Dispatcher plan and then additional sub-plans to account for far different row counts present in the VoteTypeId column.

If this scenario (and other similar scenarios) is outside the scope of the feature, query tuning folks are still going to have a whole lot of parameter sensitivity issues to deal with.

And should that be the case, I’d like to speak to the manager.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: When PSP Can Help Your Queries Go Faster

Posted on June 30, 2022June 24, 2022 by Erik Darling

Errant Hairs

I’ve used this proc as an example in the past. It’s a great parameter sniffing demo.

Why is it great? Because there’s exactly one value in the Posts table that causes an issue. It causes that issue because someone hated the idea of normalization.

The better thing to do here would be to have separate tables for questions and answers. Because we don’t have those, we end up with a weird scnenario.

In the Posts table, because questions and answers are lumped in together, there are certain traits that different types of posts can’t share:

Answers can’t have answers
Questions can’t have parent questions

There are other examples, but those are the two most obvious ones. But anyway, because of that, every Question has a ParentId of zero, and every Answer has the ParentId of the question it was posted under.

With around 6 million questions in the Posts table, that means there are around 6 million rows with a ParentId of zero, and around 11 million rows with other values.

Current Affairs

In compat level 150, if I run this procedure like so:

EXEC dbo.OptionalRecompile 
    @ParentId = 184618;

EXEC dbo.OptionalRecompile 
    @ParentId = 0;

The query plan is shared, and the second execution eats it:

From 4ms to 2 minutes is pretty bad.

2OH22

In compat level 160, if I do that exact same thing, the plans are immediately different for each execution:

This is a much better situation. Each separate plan is perfectly eligible for reuse. Neat. Great.

This is exactly the kind of query plan shenanigans (plananigans?) that should be avoided.

The only difference in the query text is the QueryVariantID:

option (PLAN PER VALUE(QueryVariantID = 2, predicate_range([StackOverflow2013].[dbo].[Posts].[ParentId] = @ParentId, 100.0, 1000000.0)))
option (PLAN PER VALUE(QueryVariantID = 3, predicate_range([StackOverflow2013].[dbo].[Posts].[ParentId] = @ParentId, 100.0, 1000000.0)))

And over in Query Store, we have these little lovelies:

SELECT qspf.* FROM sys.query_store_plan_feedback AS qspf;

+------------------+---------+------------+--------------+-----------------------+-------+-------------------+
| plan_feedback_id | plan_id | feature_id | feature_desc |     feedback_data     | state |    state_desc     |
+------------------+---------+------------+--------------+-----------------------+-------+-------------------+
|                6 |       3 |          1 | CE Feedback  | {"Feedback hints":""} |     1 | NO_RECOMMENDATION |
|                7 |       2 |          1 | CE Feedback  | {"Feedback hints":""} |     1 | NO_RECOMMENDATION |
+------------------+---------+------------+--------------+-----------------------+-------+-------------------+

Good job, SQL Server 2022.

Thanks for reading!

Going Further

SQL Server 2022 Parameter Sensitive Plan Optimization: Extended Events Related To PSP

Posted on June 29, 2022June 30, 2022 by Erik Darling

Respect Wood

There isn’t a lot interesting in these Extended Events. I was pretty disappointed with them.

The ones I looked at are in this session definition:

CREATE EVENT SESSION 
    psp
ON 
    SERVER 
ADD EVENT 
    sqlserver.parameter_sensitive_plan_optimization
(
    ACTION(sqlserver.sql_text)
),
ADD EVENT 
    sqlserver.parameter_sensitive_plan_optimization_skipped_reason
(
    ACTION(sqlserver.sql_text)
),
ADD EVENT 
    sqlserver.parameter_sensitive_plan_testing
(
    ACTION(sqlserver.sql_text)
),
ADD EVENT 
    sqlserver.query_with_parameter_sensitivity
(
    ACTION(sqlserver.sql_text)
)
ADD TARGET 
    package0.event_file
(
    SET 
        filename = N'psp'
)
WITH 
(
    MAX_MEMORY = 4096 KB,
    EVENT_RETENTION_MODE = ALLOW_SINGLE_EVENT_LOSS,
    MAX_DISPATCH_LATENCY = 1 SECONDS,
    MAX_EVENT_SIZE = 0 KB,
    MEMORY_PARTITION_MODE = NONE,
    TRACK_CAUSALITY = OFF,
    STARTUP_STATE = OFF
);

For the lazy, that’s:

parameter_sensitive_plan_optimization: This event is fired when a query uses Parameter Sensitive Plan (PSP) Optimization feature.
parameter_sensitive_plan_optimization_skipped_reason: Occurs when the parameter sensitive plan feature is skipped. Use this event to monitor the reason why parameter sensitive plan optimization is skipped
parameter_sensitive_plan_testing: Fired when parameter sensitive plan is tested.
query_with_parameter_sensitivity: This event is fired when a query is discovered to have parameter sensitivity. This telemetry will help us in identifying queries that are parameter sensitive and how skewed the columns involved in the query are.

The only one I haven’t gotten to fire yet in my testing is parameter_sensitive_plan_testing which does not break my heart.

Mythos

Most of what gets returned by those is barely useful. The one that I thought was most promising was the skipped_reason event. I was able to track map values down for that one:

+-------------------------+---------+----------------------------------+
|          name           | map_key |            map_value             |
+-------------------------+---------+----------------------------------+
| psp_skipped_reason_enum |       0 | None                             |
| psp_skipped_reason_enum |       1 | QueryVariant                     |
| psp_skipped_reason_enum |       2 | NoParameter                      |
| psp_skipped_reason_enum |       3 | InMemoryOLTP                     |
| psp_skipped_reason_enum |       4 | AutoParameterized                |
| psp_skipped_reason_enum |       5 | NonCacheable                     |
| psp_skipped_reason_enum |       6 | WithRecompileFlag                |
| psp_skipped_reason_enum |       7 | Unrecompilable                   |
| psp_skipped_reason_enum |       8 | TableVariable                    |
| psp_skipped_reason_enum |       9 | DBScopedConfigOff                |
| psp_skipped_reason_enum |      10 | QueryHint                        |
| psp_skipped_reason_enum |      11 | HasLocalVar                      |
| psp_skipped_reason_enum |      12 | QueryTextTooLarge                |
| psp_skipped_reason_enum |      13 | CursorWithPopulate               |
| psp_skipped_reason_enum |      14 | CLRModule                        |
| psp_skipped_reason_enum |      15 | Tvf                              |
| psp_skipped_reason_enum |      16 | DistributedQuery                 |
| psp_skipped_reason_enum |      17 | FullText                         |
| psp_skipped_reason_enum |      18 | OutputOrModifiedParam            |
| psp_skipped_reason_enum |      19 | UsePlan                          |
| psp_skipped_reason_enum |      20 | PDW                              |
| psp_skipped_reason_enum |      21 | Polybase                         |
| psp_skipped_reason_enum |      22 | EDC                              |
| psp_skipped_reason_enum |      23 | GQ                               |
| psp_skipped_reason_enum |      24 | DatatypesIncompat                |
| psp_skipped_reason_enum |      25 | LoadStatsFailed                  |
| psp_skipped_reason_enum |      26 | PlanGuide                        |
| psp_skipped_reason_enum |      27 | ParamSniffDisabled               |
| psp_skipped_reason_enum |      28 | NonInlinedUDF                    |
| psp_skipped_reason_enum |      29 | SkewnessThresholdNotMet          |
| psp_skipped_reason_enum |      30 | ConjunctThresholdNotMet          |
| psp_skipped_reason_enum |      31 | CompatLevelBelow160              |
| psp_skipped_reason_enum |      32 | UnsupportedStatementType         |
| psp_skipped_reason_enum |      33 | UnsupportedComparisonType        |
| psp_skipped_reason_enum |      34 | SystemDB                         |
| psp_skipped_reason_enum |      35 | UnsupportedObject                |
| psp_skipped_reason_enum |      36 | CompilationTimeThresholdExceeded |
| psp_skipped_reason_enum |      37 | Other                            |
+-------------------------+---------+----------------------------------+

Reading through that list, there are some interesting points where the feature won’t kick in. These points are sort of a ghosts of query anti-patterns past.

TableVariable
HasLocalVar
CursorWithPopulate
Tvf
ParamSniffDisabled
NonInlinedUDF

And, of course, this one gave me quite the chuckle: QueryTextTooLarge.

YOU HEAR THAT ENTITY FRAMEWORK?

Dulls

The results from my Extended Event session, and… I’ll talk through my disappointment in a moment, was not terribly spectacular.

The skipped_reason doesn’t reliably collect query text the way the other events do. That makes identifying the query that got skipped pretty difficult. There may be another action (like the stack) that captures it, but I haven’t had time to fully investigate yet.

Okay. Fine.

What do you do with this? I’m not sure yet. I don’t think I see much use in these just yet.

Hey, tomorrow’s another day. We’ll look at how the feature kicks in and fixes a parameter sniffing issue with this procedure.

Thanks for reading!