Tuesday, June 06, 2017

Having fun with maxed out uniqifiers on a non unique clustered index



I was reading the post SQL Server Non-Unique Clustered Indexes Are Evil! by the Lone DB

In that post, the person created the following table


USE boo
GO

CREATE TABLE ClusteredIndexCrash ( i TINYINT NOT NULL ) ;
CREATE CLUSTERED INDEX ClusteredIndexCrashIdx ON dbo.ClusteredIndexCrash( i ) ;


As you can see, the table has a non unique clustered index

If you were to fill this table with the same value 2147483648 times, you will get an error

So if you run this first (this will take a while believe me)


INSERT INTO dbo.ClusteredIndexCrash
SELECT TOP 1000000
    1
FROM
    syscolumns c1 , syscolumns c2 , syscolumns c3 ;

GO 2146

Then you run the following statement


INSERT INTO dbo.ClusteredIndexCrash
SELECT TOP 1000000
    1
FROM
    syscolumns c1 , syscolumns c2 , syscolumns c3 ;


You would get the following error..straight from the beast himself apparently

Msg 666, Level 16, State 2, Line 1
The maximum system-generated unique value for a duplicate group was exceeded for index with partition ID 72057594039173120. Dropping and re-creating the index may resolve this; otherwise, use another clustering key.


I will be using DBCC PAGE and DBCC IND in this blog post, if you want to learn how to use these yourself, take a look at How to use DBCC PAGE


I decided to look in the pages of SQL Server itself. I grabbed all the pages by running DBCC IND


DBCC IND ('boo','ClusteredIndexCrash',  1);


That returned 4806619  rows, these are all the pages that exist for this table

Here is part of the output that matters, the highlighted row where NextPageFID is 0 is what we want, this is the last page.



Click on the image for a larger view


Then I grabbed the last page id from that result set and looked at the page by executing these commands


DBCC TRACEON (3604);
GO
DBCC PAGE (boo, 1, 4810967, 3) with TABLERESULTS;



Here is part of the output in text, I highlighted the uniqifier for you


PAGE HEADER: Slot 158 Offset 0xa40 Length 16 Record Type PRIMARY_RECORD

PAGE HEADER: Slot 158 Offset 0xa40 Length 16 Record Attributes NULL_BITMAP VARIABLE_COLUMNS

PAGE HEADER: Slot 158 Offset 0xa40 Length 16 Record Size 16

Slot 158 Offset 0xa40 Length 16 Memory Dump @0x00000000E832AA40 0000000000000000: 30000500 01020000 01001000 fdffff7f 0...........ýÿÿ.

Slot 158 Offset 0xa40 Length 16 Slot 158 Column 0 Offset 0xc Length 4 Length (physical) 4 UNIQUIFIER 2147483645

Slot 158 Offset 0xa40 Length 16 Slot 158 Column 1 Offset 0x4 Length 1 Length (physical) 1 i 1

Slot 158 Offset 0xa40 Length 16 Slot 158 Offset 0x0 Length 0 Length (physical) 0 KeyHashValue (5eeffbfe6491)

PAGE HEADER: Slot 159 Offset 0xa50 Length 16 Record Type PRIMARY_RECORD

PAGE HEADER: Slot 159 Offset 0xa50 Length 16 Record Attributes NULL_BITMAP VARIABLE_COLUMNS

PAGE HEADER: Slot 159 Offset 0xa50 Length 16 Record Size 16

Slot 159 Offset 0xa50 Length 16 Memory Dump @0x00000000E832AA50 0000000000000000: 30000500 01020000 01001000 feffff7f 0...........þÿÿ.

Slot 159 Offset 0xa50 Length 16 Slot 159 Column 0 Offset 0xc Length 4 Length (physical) 4 UNIQUIFIER 2147483646

Slot 159 Offset 0xa50 Length 16 Slot 159 Column 1 Offset 0x4 Length 1 Length (physical) 1 i 1

Slot 159 Offset 0xa50 Length 16 Slot 159 Offset 0x0 Length 0 Length (physical) 0 KeyHashValue (bedbd571a02d)




Here is the same output as an image from SSMS




Click on the image for a larger view



I decided to play around with this now that I had the table maxed out

What would happen if I deleted a bunch of rows? Would I still get the same error? I deleted 100 rows

DELETE TOP (100)
FROM  ClusteredIndexCrash


Now I tried to insert 1 row..

INSERT INTO dbo.ClusteredIndexCrash
SELECT 1

Msg 666, Level 16, State 2, Line 1
The maximum system-generated unique value for a duplicate group was exceeded for index with partition ID 72057594039173120. Dropping and re-creating the index may resolve this; otherwise, use another clustering key.

So same issue, doesn't reuse uniqifiers


Would an index rebuild fix this?


ALTER INDEX ClusteredIndexCrashIdx ON ClusteredIndexCrash REBUILD


-- 20 minutes later.......



INSERT INTO dbo.ClusteredIndexCrash
SELECT 1


Msg 666, Level 16, State 2, Line 1
The maximum system-generated unique value for a duplicate group was exceeded for index with partition ID 72057594039566336. Dropping and re-creating the index may resolve this; otherwise, use another clustering key.



Nope, nothing changed


What about a truncate?


TRUNCATE TABLE ClusteredIndexCrash


Now when I run DBCC IND...


DBCC IND ('boo','ClusteredIndexCrash',  1);

This returns nothing.. that is fine..after all this table is empty


Let's try adding a row


INSERT INTO dbo.ClusteredIndexCrash
SELECT 1


No problem, error is gone


I can now again use DBCC IND like before


DBCC IND ('boo','ClusteredIndexCrash', 1);

But since I know I only have 1 row now, I can also use the sys.fn_PhysLocCracker function together with the %%physloc%% virtual column, it will actually give me the page id the row is located on

Here is the query

SELECT top 10 *,%%physloc%% AS PhysicalLocationOfRow
FROM ClusteredIndexCrash
CROSS APPLY sys.fn_PhysLocCracker(%%physloc%%) AS FPLC

This is the output from that query

i file_id page_id slot_id PhysicalLocationOfRow
1 1 7668633 0 0x9903750001000000


Running the DBCC PAGE command again with the pageid from the query above


DBCC TRACEON (3604);
GO
DBCC PAGE (boo, 1, 7668633, 3) with TABLERESULTS;



And here is the output from DBCC PAGE



PAGE HEADER: Slot 0 Offset 0x60 Length 9 Record Type PRIMARY_RECORD

PAGE HEADER: Slot 0 Offset 0x60 Length 9 Record Attributes NULL_BITMAP

PAGE HEADER: Slot 0 Offset 0x60 Length 9 Record Size 9

Slot 0 Offset 0x60 Length 9 Memory Dump @0x00000000C41EA060 0000000000000000: 10000500 01020000 01 .........

Slot 0 Offset 0x60 Length 9 Slot 0 Column 0 Offset 0x0 Length 4 Length (physical) 0 UNIQUIFIER 0

Slot 0 Offset 0x60 Length 9 Slot 0 Column 1 Offset 0x4 Length 1 Length (physical) 1 i 1

Slot 0 Offset 0x60 Length 9 Slot 0 Offset 0x0 Length 0 Length (physical) 0 KeyHashValue (fa469958a30a)


Here it is also as an image




Click on the image for a larger view


As you can see the uniqifier is now 0

So it looks like the uniqifier behaves like an identity property per value, a truncate will reseed it back to 0 but when you delete values uniqifiers are not reused

So maybe we would need a DBCC CHECKUNIQIFIER command  :-)


So how important is this..will you ever run into this?  I can see this becoming an issue if you have a staging table which you don't drop or truncate but you constantly insert and delete data from it

Maybe you should drop the table once in a while?


Have you ever maxed out the uniqifier?

Saturday, May 13, 2017

All the Build 2017 data related sessions in 1 spot, SQL Server, CosmosDB, Azure SQL DB and more




Are you bored this weekend? No problem, I have got you covered, I have put together a page with all the data related sessions from the build 2017 conference. I have embedded the videos and also added links to the page at the channel 9 site. I have also added links to the presentations and source code if it was available.  I have arranged this in 4 categories:

SQL Server
Azure SQL DB
CosmosDB
The rest of the sessions


This way the sessions are not all intermingled

SQL Server


Securing your app data with Microsoft SQL Server 2017 
by Tommy Mullaney

Join us for an overview of strategy and best practices to build secure applications using SQL Server. SQL Server offers a versatile toolset of security capabilities, which enable you to protect sensitive data and meet stringent compliance requirements. You’ll leave with an understanding of how these capabilities fit together to protect your application as a whole, including the new security enhancements available in the latest version of SQL Server.




Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4018



Evolving SQL workloads from software to SaaS: Data for devs immersion
by Tara Shankar Jana

In this modern era; applications must recover from user errors efficiently, must be optimized for transactional performance, provide real-time insights into app transactions, must be secure and complaint, and support seamless integration to cloud, hybrid scenarios and other data types/systems. These applications must also learn and adapt to deliver more powerful experiences using cloud-based data services, applications, and intelligence. In this session we use a scenario-based hands-on lab program (immersion), to offer you a cohesive and consistent experience to build modern on-premises/cloud based applications with intelligence. Build hack-free, high performance, scalable applications using Microsoft SQL Server, Microsoft ML and Azure data services.



Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4013



Advanced SQL Server on Linux (Docker) via Entity Framework (EF)
by Tara Shankar Jana

It’s SQL Server as you already know it today, but it’s now native to Linux. It works with your data, your favorite tools, application frameworks (Entity Framework) and programming languages. In this session, we pull the code for an application running in cloud to a Linux-based machine. We test and enhance the application for performance (using EF, in-memory) and security against SQL vNext (Linux). Then we re-deploy the changes using DevOps to Azure with just a flip of d, show how the application works across platforms, on-premises, and in the cloud. Best part, all of it will be done in ten minutes.



Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4001


SQL Unplugged
by Scott Klein, Rohan Kumar

In this Channel 9 Live session we'll discuss all things SQL



Link to Build site: https://channel9.msdn.com/Events/Build/2017/C9L20


Serving AI with data: How to use R, Python, and machine learning with Microsoft SQL Server 2017
by Umachandar Jayachandran, Nellie Gustafsson

In this session, learn how to use R and Python to perform statistical and machine learning tasks in SQL Server 2017. Also, learn how to operationalize your R or Python scripts using the SQL Server integration.




download:Slides View Slides Online




Microsoft SQL Server 2017 in the modern datacenter: how to run on Linux, Docker, OpenShift, and Kubernetes
by Travis Wright, Tobias Ternstrom

With the on-premises data center evolving into a private cloud, the fast pace of innovation in public clouds, and the strong interest in container technology it begs the question: How do you run a database in this new environment? In this session we examine what it means that SQL Server is evolving to support both Windows and Linux, including running inside a container on both platforms. Should you run SQL Server inside a Docker container? How does it play with orchestration technologies like OpenShift and Kubernetes? The brief answer is that we believe SQL Server should be ubiquitous and run in our customers environment of choice. Come to the session to hear the long answer with a bunch of demos!



Link to Build site: https://channel9.msdn.com/Events/Build/2017/B8080


How to do Predictive Modelling using R and SQL Server ML Services
4 days ago  by Umachandar Jayachandran, Nellie Gustafsson

Learn how to use R scripts from T-SQL to perform training and scoring and leverage parallelism and streaming capabilities to get better performance.



Link to Build site: https://channel9.msdn.com/Events/Build/2017/T6070


Built-in machine learning in Microsoft SQL Server 2017 with Python
by Sumit Kumar

Machine learning services in SQL Server 2017 provides Python support for in-database machine learning, now. In this session we show the basics of how to run Python code in SQL Server. We then discuss how any app that can talk to SQL Server can get intelligence from machine learning models running in SQL Server. We showcase an app that uses a Python-based deep learning model built and deployed in SQL Server. The model leverages an open source deep learning framework running with SQL Server and utilizes GPU for boosting model training performance.




Link to Build site: https://channel9.msdn.com/Events/Build/2017/T6067



Modernize your database development lifecycle with SQL Server Data Tools in Visual Studio
2 days ago  by Tara Raj

Learn how SQL Server Data Tools (SSDT) turns Visual Studio into a powerful environment for database development. Easily build, debug, maintain, and refactor databases inside Visual Studio with a declarative model that spans all the phases of database development and easily enables continuous integration and deployment for your databases. Work offline with a database project, or work directly with a connected database instance in Azure SQL Database, Azure SQL Data Warehouse, and SQL Server running on Windows, Linux, or Docker, on-premises or in any cloud.


Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4009

Azure SQL DB

Get to the cloud faster with Azure SQLDB Managed Instance and Database Migration Service
by Lindsey Allen, Harini Gupta


A new, expanded Azure SQL Database offers fully-managed server instance with greater compatibility with SQL Server application features and more, and the ability to move hundreds of databases at once using the Database Migration Service. It provides security isolation with Azure Virtual Network, along with built-in HADR, built-in intelligent performance tuning, and intelligent security services that are already available in Azure SQL Database. To reduce the friction of transitioning your relational database estate to public cloud, expanded Azure SQL Database provides SQL Server application compatibility including commonly used cross database references using three-part names, CLR, SQL Agent, Transactional Replication, Change Data Capture, Service Broker, and a lot more. We showcase a five-step seamless migration experience from your on-premises SQL Server instances to expanded Azure SQL Database using the newly announced Azure Database Migration Service.



Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4008


Migrating Oracle database to the Azure SQL DB with Database Migration Service
by Shamik Ghosh, Alexander Ivanov

In this virtual session, you learn how to assess, remediate and migrate the schema, data, and server artifacts from on-premises Oracle instances to Azure SQL DB. We walk you through an end-to-end experience using Azure data migration tools and services to help solve your database migration needs.




Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4182 

download: Slides View Slides Online



How to build global-scale applications with Microsoft Azure SQL Database
by Rohan Kumar, Bob Strudwick

Join us in this session to learn how to build a global-scale IoT application by walking through a sample application and real-world customer case study. Gain insight on building an IoT application on a fully managed Azure database as a service, with built-in intelligence for performance, cost and security, dynamic scaling and hassle-free HADR, and infrastructure maintenance. We show you how to use Azure SQL Database enterprise grade in-memory engines to handle high volume and high concurrent transactions; while running real-time analytics queries on the same database. New features are released in Azure SQL Database in fast cadence; we show you the new Graph processing in the context of an IoT scenario, Adaptive Query Processing, predictive analytics, and the first in market built-in homomorphic data encryption feature for securing data at rest and in motion. It will be a fun learning hour.


Link to Build site: https://channel9.msdn.com/Events/Build/2017/B8018


Design Patterns for SaaS applications on Azure SQL Database
by Julie Strauss, Bill Gibson


Experience the power of building multi-tenant SaaS applications on Azure SQL Database, Microsoft’s fully managed database as a service platform: Using a sample SaaS application, we walk through a series of SaaS-focused design and management patterns that have been distilled from work with a multitude of customers. Patterns spans from multi-tenant provisioning, schema management, performance monitoring and management to operational analytics. The code for the sample application, plus management scripts, ARM templates and tutorials, will be available for download in an easy-to-explore “SaaS-in-a-Box” package, enabling you to jump-start your own SaaS application.


Link to Build site: https://channel9.msdn.com/Events/Build/2017/T6025

CosmosDB


CosmosDB
by Seth Juarez, Rimma Nehme

In this session we meet with Rimma Nehme and Seth Juarez to discuss some of the latest Build 2017 announcements around CosmosDB



Link to Build site: https://channel9.msdn.com/Events/Build/2017/C9L08


Azure Cosmos DB: Build planet scale mobile apps in minutes
by Kirill Gavrylyuk

This session has no description, but the video is below



Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4012


Azure Cosmos DB: API for MongoDB
by Andrew Hoh

Azure Cosmos DB (formerly known as Azure DocumentDB) natively supports multiple APIs; one of which is the API for MongoDB. Use existing code, applications, drivers, and tools to work with Azure Cosmos DB. Benefit from the fully managed and scalable Azure database, while continuing to use familiar skills and tools for MongoDB. Come and watch this video to learn about the feature and how to migrate to this cosmic-scale database, Azure Cosmos DB.




Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4011


Azure Cosmos DB: NoSQL capabilities everyone should know about
by Aravind Ramachandran

Microsoft Azure provides a fully managed NoSQL database service built for fast and predictable performance, high availability, elastic scaling, global distribution, and ease of development. As a schema-free NoSQL database, the service provides rich and familiar SQL query capabilities with consistent low latencies on JSON data - ensuring that 99% of your reads are served under 10 milliseconds and 99% of your writes are served under 15 milliseconds. These unique benefits make it a great fit for web, mobile, gaming, IoT, AI, and many other applications that need seamless scale and global replication. Come and learn about the NoSQL capabilities in Azure Cosmos DB that every developer should know about.



Link to Build site: https://channel9.msdn.com/Events/Build/2017/T6058



A lap around Azure HDInsight and Cosmos DB Open Source Analytics + NoSQL
by Andrew Liu, Raghav Mohan


Recently, we released the Spark Connector for our distributed NoSQL service – Azure Cosmos DB (formerly known as Azure DocumentDB). By connecting Apache Spark running on top Azure HDInsight to Azure Cosmos DB, you can accelerate your ability to solve fast-moving data science problems and machine learning. The Spark to Azure Cosmos DB connector efficiently exploits the native Cosmos DB managed indexes and enables updateable columns when performing analytics, push-down predicate filtering against fast-changing globally-distributed data, ranging from IoT, data science, and analytics scenarios. Come learn how you can perform blazing fast planet-scale data processing with Azure Cosmos DB and HDInsight.


Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4010


Data lakes, U-SQL, Azure DBMS  and the rest


How to get started with the new Azure DBMS for PostgreSQL.
by Eric Spear, Irakliy Khaburzaniya, Sunil Kamath

Join this session to learn about Microsoft’s latest announcement of managed PostgreSQL database on Azure. In this breakout, we will learn from two early adopting customers, how they've leveraged this latest database service to innovate faster. We will learn from their experience of using the managed PostgreSQL service, including migrating to the service, and discuss next steps in their application journey. We will walk through some of the key service features and discuss how you as a developer can migrate your existing applications or develop new applications that use this managed PostgreSQL in Azure. If you’re a developer with applications that use PostgreSQL today, whether on-premises or cloud, and want to learn about how this new managed service can help, this session is for you!



Link to Build site: https://channel9.msdn.com/Events/Build/2017/B8046
download: Slides View Slides Online



How to get started with the new Azure DBMS for MySQL.
by Gebi Liang, Jason M. Anderson, Matt Williams


Join this session to learn about Microsoft’s managed MySQL offering in Azure. We’ll walk through Microsoft’s strategy for supporting Open-Source database systems in Azure and what it means to you as a developer as you look to develop or deploy applications that use MySQL in Azure. An overview of the architecture of the service along with how Azure Database for MySQL is integrated with other Azure Services such as Web Apps will also be discussed and demo’d. If you’re a developer with applications using MySQL today, whether on-prem or already in Azure today, this session is for you!



Link to Build site: https://channel9.msdn.com/Events/Build/2017/B8045
download: Slides View Slides Online Source Code



How to serve AI with data: The future of the Microsoft Data Platform
by Joseph Sirosh, Alex Miller

The cloud is truly becoming the “brain” for our connected planet. You’re not just running algorithms in the cloud, rather you’re connecting that with data from sensors from around the world. By bringing data into the cloud, you can integrate all of the information, apply machine learning and AI on top of that, and deliver apps that are continuously learning and evolving in the cloud. Consequently, the devices and applications connected to cloud are learning and becoming increasingly intelligent. Please join us on a journey through the core new patterns that are emerging as we bring advanced intelligence, the cloud, IoT, and big data together.




Link to Build site: https://channel9.msdn.com/Events/Build/2017/B8081


How to run AI at Petabyte Scale with cognitive functions in the Azure Data Lake
by Wee Hyong Tok

In this session, learn how you can use Azure Data Lake (ADL) for doing Big Cognition on raw customer support data. Learn how you can use ADL to perform key phrase extraction, sentiment analysis, and how you can use R/Python scripts for support volume forecasting. In addition, learn how you can use federated queries in ADL with Azure SQL Database. Discover how you can pull all these insights into an Azure SQL Data Warehouse, and using Azure Analysis Services to enable interactive analysis of the processed data. Join us for this exciting session as we show how you can develop intelligent applications by using the insights derived from processing massive Petabyte-scale datasets.


Link to Build site: https://channel9.msdn.com/Events/Build/2017/B8065
download: Slides View Slides Online Source Code



Image processing at scale using U-SQL in Azure Data Lake
by Saveen Reddy

Making use of cognitive capabilities such as Image OCR or Sentiment Analysis of text is straightforward with small datasets of a few terabytes. But, at the scale of hundreds of terabytes or even a petabyte, you need a different approach to that can massively scale out *AND* be simple to build. Azure Data Lake offers a straightforward way of programming using .NET code against these massive Petabyte-scale datasets without the need to become a deep expert in distributed computing, big data technologies, or machine learning.Link to Build site:


Link to Build site: https://channel9.msdn.com/Events/Build/2017/T6040



Lift and shift any runtime into U-SQL for large scale processing
by Michael Rys

One of U-SQL’s major strengths is to scale out user code over your large data lake assets. This presentation first shows how we have used the U-SQL extensibility model to give you access to Python and R and then shows how to run Scala from within U-SQL.


Link to Build site: https://channel9.msdn.com/Events/Build/2017/P4004


There you have have a bunch of data related sessions for your viewing pleasure




Wednesday, May 10, 2017

Not sure that I like the message from Online Resumable Index Rebuild in SQL Server 2017


I was messing around with the Online Resumable Index Rebuild, this is new in CTP 2 of SQL Server 2017. I don't know that I like the output from an resumable index rebuild

Let's take a look at what I did
First I created this table

CREATE TABLE dbo.TestIndexRebuild(
 name nvarchar(35) NULL,
 number int NOT NULL,
 type nchar(3) NOT NULL,
 low int NULL,
 high int NULL,
 status int NULL,
 somebigchar char(2000)
) ON [PRIMARY]

CREATE CLUSTERED INDEX CI_TestIndexRebuild ON TestIndexRebuild(name,number)

I made the table wide by adding a 2000 character column, I then added a clustered index to the table

I then pumped in a bunch of data
INSERT INTO TestIndexRebuild
SELECT *,REPLICATE('A',2000)  
FROM master..spt_values
GO 500


I then executed the following
ALTER INDEX CI_TestIndexRebuild on TestIndexRebuild 
REBUILD WITH (ONLINE=ON, RESUMABLE=ON, MAX_DURATION=1) 

Here is the output I got after 1 minute

Msg 3643, Level 16, State 1, Line 19
The operation elapsed time exceeded the maximum time specified for this operation. The execution has been stopped.
Msg 596, Level 21, State 1, Line 18
Cannot continue the execution because the session is in the kill state.
Msg 0, Level 20, State 0, Line 18
A severe error occurred on the current command.  The results, if any, should be discarded.

I don't really like that, index rebuild has been paused because elapsed time has exceeded the maximum time displayed in black instead of red would have been fine with me. I don't need to see that the session is in a kill state or that a severe error occurred

I then executed the same command again

ALTER INDEX CI_TestIndexRebuild on TestIndexRebuild 
REBUILD WITH (ONLINE=ON, RESUMABLE=ON, MAX_DURATION=1) 

Here is the output

Warning: An existing resumable operation with the same options was identified for the same index on 'TestIndexRebuild'. The existing operation will be resumed instead.

While that was running in a second window, I executed the following

ALTER INDEX CI_TestIndexRebuild on TestIndexRebuild  ABORT 

All you get as output is

Command(s) completed successfully.

However, in the widow where you executed the resumable index rebuild you get this

Msg 1219, Level 16, State 1, Line 1
Your session has been disconnected because of a high priority DDL operation.
Msg 596, Level 21, State 1, Line 0
Cannot continue the execution because the session is in the kill state.
Msg 0, Level 20, State 0, Line 0
A severe error occurred on the current command.  The results, if any, should be discarded.

If you execute the ABORT again while the index is not being rebuilt, you get

Msg 10638, Level 16, State 2, Line 1
ALTER INDEX 'ABORT' failed. There is no pending resumable index operation for the index 'CI_TestIndexRebuild' on 'TestIndexRebuild'.

That makes sense


With a PAUSE Same thing happens as with ABORT, when you pause the index rebuild


ALTER INDEX CI_TestIndexRebuild on TestIndexRebuild  PAUSE

All you get as output is

Command(s) completed successfully

But you get those other messages in the original window


You can also execute a PAUSE followed by an ABORT, you will only get one set of messages, no error is displayed in the window where you executed the code below



ALTER INDEX CI_TestIndexRebuild on TestIndexRebuild  PAUSE 
GO
ALTER INDEX CI_TestIndexRebuild on TestIndexRebuild  ABORT 





I think the kill state and severe error occurred is a little over the top.
What is your opinion?


Now having said all that, I do like the resumable index rebuilds, it pretty much mimics the defragment/reorganize functionality. It continues from where it was when if was running last

There are some more things you can specify, for example, here is a sample command

ALTER INDEX test_idx on test_table RESUME WITH (MAXDOP=2, MAX_DURATION= 240 MINUTES, WAIT_AT_LOW_PRIORITY (MAX_DURATION=10, ABORT_AFTER_WAIT=BLOCKERS)) ;


Read more about this in section J of ALTER INDEX


Tuesday, May 09, 2017

SQL Server 2017: SQL Graph



With the CTP2 build of SQL Server 2017, you now have the ability to use SQL Graph

I decided to play around with it a little, I based the code here on the sample I found in Books On Line. To follow along, first create this database, you have to be on SQL Server 2017 CTP2 or higher in order for the code to work



CREATE DATABASE GraphDBTest;
GO

USE  GraphDBTest;
GO


In SQL Graph, you have node tables and edge table, here is the description for both tables from Books On Line


A node table represents an entity in a graph schema. Every time a node table is created, along with the user defined columns, an implicit $node_id column is created, which uniquely identifies a given node in the database. The values in $node_id are automatically generated and are a combination of object_id of that node table and an internally generated bigint value

It is recommended that users create a unique constraint or index on the $node_id column at the time of creation of node table, but if one is not created, a default unique, non-clustered index is automatically created.



An edge table represents a relationship in a graph. Edges are always directed and connect two nodes. An edge table enables users to model many-to-many relationships in the graph. An edge table may or may not have any user defined attributes in it. Every time an edge table is created, along with the user defined attributes, three implicit columns are created in the edge table:

Column nameDescription
$edge_idUniquely identifies a given edge in the database. It is a generated column and the value is a combination of object_id of the edge table and a internally generated bigint value. However, when the $edge_id column is selected, a computed value in the form of a JSON string is displayed. $edge_id is a pseudo-column, that maps to an internal name with hex string in it. When you select $edge_id from the table, the column name will appear as $edge_id_\<hex_string>. Using pseudo-column names in quereis is the recommended way of querying the internal $edge_id column and using internal name with hex string should be avoided.
$from_idStores the $node_id of the node, from where the edge originates.
$to_idStores the $node_id of the node, at which the edge terminates.


Similar to the $node_id column, it is recommended that users create a unique index or constraint on the $edge_id column at the time of creation of the edge table, but if one is not created, a default unique, non-clustered index is automatically created on this column. It is also recommended, for OLTP scenarios, that users create an index on ($from_id, $to_id) columns, for faster lookups in the direction of the edge.


Here is what a node table looks like, the only difference is the AS NODE at the end

CREATE TABLE Person (
  ID INT PRIMARY KEY,
  name VARCHAR(100)
) AS NODE;

Inserting into a node table is same as inserting into a regular table


INSERT INTO Person VALUES (1,'Denis');
INSERT INTO Person VALUES (2,'Shidhar');
INSERT INTO Person VALUES (3,'Geoff');
INSERT INTO Person VALUES (4,'Kevin');
INSERT INTO Person VALUES (5,'Liz');
INSERT INTO Person VALUES (6,'Melanie');

Let's see what the output looks like

SELECT * FROM Person


$node_id_A0703BD81C174451BEC7688009010A5B ID name
{"type":"node","schema":"dbo","table":"Person","id":0} 1 Denis
{"type":"node","schema":"dbo","table":"Person","id":1} 2 Shidhar
{"type":"node","schema":"dbo","table":"Person","id":2} 3 Geoff
{"type":"node","schema":"dbo","table":"Person","id":3} 4 Kevin
{"type":"node","schema":"dbo","table":"Person","id":4} 5 Liz
{"type":"node","schema":"dbo","table":"Person","id":5} 6 Melanie

As you can see, there is JSON in the $node_id column

The sys.tables view has 2 new columns, is_edge and is_node. If we query the sys.tables view now for the Person table we just created, you will see that is_node = 1

SELECT t.is_edge,t.is_node,*
FROM sys.tables t
WHERE name = 'Person'

Here is the result

is_edgeis_nodenameobject_idschema_idtype_desccreate_date
01Person9015782501USER_TABLE2017-05-09 08:55:30.513


Now we will add another table, this table will have some data about DB products



CREATE TABLE DBProduct (
  ID INT NOT NULL, 
  name VARCHAR(100)
) AS NODE;



INSERT INTO DBProduct VALUES (1,'SQL Server');
INSERT INTO DBProduct VALUES (2,'Cassandra');
INSERT INTO DBProduct VALUES (3,'Oracle');
INSERT INTO DBProduct VALUES (4,'FoundationDB');
INSERT INTO DBProduct VALUES (5,'MongoDB');
INSERT INTO DBProduct VALUES (6,'Vertica');
INSERT INTO DBProduct VALUES (7,'Volt');



And now it is time for our edge tables, we will have 2 of them, a likes table and a friendOf table

CREATE TABLE likes (rating INT) AS EDGE;
CREATE TABLE friendOf AS EDGE;


To insert into the likes table, we have to say which person likes what product. The person will go into the $from_id column, the DBProduct will got into the $to_id column. You $Node_id to grab the id to populate this table


The node_id for Person with an id of 1 will go into the $from_id column, the node_id for DBProduct with an id of 1 will go into the $to_id column. That looks like this

((SELECT $node_id FROM Person WHERE id = 1), 
       (SELECT $node_id FROM DBProduct WHERE id = 1),5);

Here are the insert statements to populate this table


INSERT INTO likes VALUES ((SELECT $node_id FROM Person WHERE id = 1), 
       (SELECT $node_id FROM DBProduct WHERE id = 1),5);
INSERT INTO likes VALUES ((SELECT $node_id FROM Person WHERE id = 1), 
       (SELECT $node_id FROM DBProduct WHERE id = 2),15);
INSERT INTO likes VALUES ((SELECT $node_id FROM Person WHERE id = 2), 
      (SELECT $node_id FROM DBProduct WHERE id = 1),6);
INSERT INTO likes VALUES ((SELECT $node_id FROM Person WHERE id = 3), 
      (SELECT $node_id FROM DBProduct WHERE id = 3),7);
INSERT INTO likes VALUES ((SELECT $node_id FROM Person WHERE id = 4), 
      (SELECT $node_id FROM DBProduct WHERE id = 4),8);
INSERT INTO likes VALUES ((SELECT $node_id FROM Person WHERE id = 5), 
      (SELECT $node_id FROM DBProduct WHERE id = 5),9);


Let's look at what it looks like in the likes edge table

SELECT * FROM likes

Here is what it looks like, as you can see it is a bunch of JSON.. (click on the image for a bigger version)


 

 Now we will also insert some data into the friendOf table, it is the same as with the likes table, lookup the node ids and insert the data

INSERT INTO friendof VALUES ((SELECT $NODE_ID FROM person WHERE ID = 1), 
    (SELECT $NODE_ID FROM person WHERE ID = 2));
INSERT INTO friendof VALUES ((SELECT $NODE_ID FROM person WHERE ID = 1), 
    (SELECT $NODE_ID FROM person WHERE ID = 5));

INSERT INTO friendof VALUES ((SELECT $NODE_ID FROM person WHERE ID = 2), 
    (SELECT $NODE_ID FROM person WHERE ID = 3));
INSERT INTO friendof VALUES ((SELECT $NODE_ID FROM person WHERE ID = 3), 
    (SELECT $NODE_ID FROM person WHERE ID = 1));
INSERT INTO friendof VALUES ((SELECT $NODE_ID FROM person WHERE ID = 4), 
    (SELECT $NODE_ID FROM person WHERE ID = 2));
INSERT INTO friendof VALUES ((SELECT $NODE_ID FROM person WHERE ID = 5), 
    (SELECT $NODE_ID FROM person WHERE ID = 4));


Now it is time to run some queries
This query will return all the products that Denis likes

-- Find DBProducts that Denis likes
SELECT DBProduct.name
FROM Person, likes, DBProduct
WHERE MATCH (Person-(likes)->DBProduct)
AND Person.name = 'Denis';

name
-------------
SQL Server
Cassandra

Not surprisingly, SQL Server is in that list

What about my friends? What DB products do they like?

-- Find DBProducts that Denis' friends like
SELECT DBProduct.name as DBName,person2.name  
FROM Person person1, Person person2, likes, friendOf, DBProduct
WHERE MATCH(person1-(friendOf)->person2-(likes)->DBProduct)
AND person1.name='Denis';


DBName                  name
---------------         ------------
SQL Server              Shidhar
MongoDB                 Liz


Okay, SQL Server and MongoDB, I can live with that

What about Liz, what does her friend Kevin like?

SELECT DBProduct.name as DBName,person2.name 
FROM Person person1, Person person2, likes, friendOf, DBProduct
WHERE MATCH(person1-(friendOf)->person2-(likes)->DBProduct)
AND person1.name='Liz';


DBName             name  
------ -------
FoundationDB     Kevin

FoundationDB? Didn't Apple buy that and then shut down the downloads?


So what product is there that both friends like?

To do that we specify that person1 is a friendOf person 2
and both person1 and person2 like the same product. Since you cannot use the same table twice, you need to use an alias, in this case I have used l2 as an alias for the likes table


SELECT DBProduct.name as DBName,person1.name,person2.name 
FROM Person person1, Person person2, likes, friendOf, DBProduct, likes as l2
WHERE MATCH(person1-(friendOf)->person2-(likes)->DBProduct 
  AND person1-(l2)->DBProduct)


DBName        name       name
-----------
SQL Server    Denis      Shidhar

What a surprise, can you believe it, it is SQL Server!!



Let's take a look at some meta data, this time, we are adding some columns from the sys.columns view, I have added graph_type_desc and is_edge

select t.name, c.name, c.graph_type_desc,t.is_edge,t.is_node 
from sys.columns c
join sys.tables t on c.object_id = t.object_id
where graph_type is not null
order by t.name, c.name

Here is the output from that query

name name graph_type_desc is_edge is_node
DBProduct $node_id_4C8D1144EF7745FD8BACDB7178353A65 GRAPH_ID_COMPUTED 0 1
DBProduct graph_id_8FA26FC01AE94CF9921364ECEFB71621 GRAPH_ID 0 1
friendOf $edge_id_FBBA05CB4E0F49E7919EC346DBB21F92 GRAPH_ID_COMPUTED 1 0
friendOf $from_id_143FE9B1B3E942EB99FF8396B292CD19 GRAPH_FROM_ID_COMPUTED 1 0
friendOf $to_id_DFF0DCFA07FE45E3A31E1925F5E6F886 GRAPH_TO_ID_COMPUTED 1 0
friendOf from_id_71FB1C6C6D82428889276A07A9AF32E4 GRAPH_FROM_ID 1 0
friendOf from_obj_id_4310B8C357D641EC990E91329687BDAA GRAPH_FROM_OBJ_ID 1 0
friendOf graph_id_3722B217CA404686BE44855E883AA6FA GRAPH_ID 1 0
friendOf to_id_D01976015F664F91A0A29819D79EE6A5 GRAPH_TO_ID 1 0
friendOf to_obj_id_8C26534B00DF47FBA87FBED6D14C3201 GRAPH_TO_OBJ_ID 1 0
likes $edge_id_CA70E11B83A44BD99DAD3B14EFCC2E08 GRAPH_ID_COMPUTED 1 0
likes $from_id_FF1F089BA1224A00A1C94F68415C6B08 GRAPH_FROM_ID_COMPUTED 1 0
likes $to_id_7728587E08684D94961B4F2234702B87 GRAPH_TO_ID_COMPUTED 1 0
likes from_id_ACD968F7815C4A3A9B5B6C5506235EDE GRAPH_FROM_ID 1 0
likes from_obj_id_185C6B5EB64F4C0A8E3E1CF65092F706 GRAPH_FROM_OBJ_ID 1 0
likes graph_id_16575BE2C0764232B5B440524166DF15 GRAPH_ID 1 0
likes to_id_D164826C35374D3E98EDBC3FA77457B3 GRAPH_TO_ID 1 0
likes to_obj_id_373354D4D7014EC89771C48F961561BB GRAPH_TO_OBJ_ID 1 0
Person $node_id_A0703BD81C174451BEC7688009010A5B GRAPH_ID_COMPUTED 0 1
Person graph_id_40FFD8418C1343639EA62CA66B6CE414 GRAPH_ID 0 1



You might have noticed that SQL Graph queries use old style joins
So instead of this

SELECT DBProduct.name as DBName,person1.name,person2.name 
FROM Person person1, Person person2, likes, friendOf, DBProduct, likes as l2

I can do this right?

SELECT DBProduct.name as DBName,person1.name,person2.name 
FROM Person person1
CROSS JOIN Person person2
CROSS JOIN likes
CROSS JOIN friendOf
CROSS JOIN DBProduct
CROSS JOIN likes as l2


Yes that works, both queries return 54432 rows

Will it work with the WHERE clause?

SELECT DBProduct.name as DBName,person1.name,person2.name 
FROM Person person1
CROSS JOIN Person person2
CROSS JOIN likes
CROSS JOIN friendOf
CROSS JOIN DBProduct
CROSS JOIN likes as l2
WHERE MATCH(person1-(friendOf)->person2-(likes)->DBProduct 
  AND person1-(l2)->DBProduct)


Msg 13920, Level 16, State 1, Line 164
Identifier 'person1' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 164
Identifier 'friendOf' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 164
Identifier 'person2' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 164
Identifier 'person2' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 164
Identifier 'likes' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 164
Identifier 'DBProduct' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 165
Identifier 'person1' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 165
Identifier 'l2' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.
Msg 13920, Level 16, State 1, Line 165
Identifier 'DBProduct' in a MATCH clause is used with a JOIN clause or APPLY operator. JOIN and APPLY are not supported with MATCH clauses.


Boooo ... that is a big fat nope



Also keep the following in mind
There are certain limitations on node and edge tables in this release:
  • Local or global temporary tables cannot be node or edge tables.
  • Table types and table variables cannot be declared as a node or edge table.
  • Node and edge tables cannot be created as system-versioned temporal tables.
  • Node and edge tables cannot be memory optimized tables.
  • Users cannot update the $from_id and $to_id columns of an edge using UPDATE statement. To update the nodes that an edge connects, users will have to insert the new edge pointing to new nodes and delete the previous one.


This was just a first look at Graph SQL, I will have to play around with it a little more to get more familiar with it and see if there are any use cases for me


In the meantime, visit Books On Line and start playing around with SQL Graph: https://docs.microsoft.com/en-us/sql/relational-databases/graphs/sql-graph-overview