Search Results

Search found 31421 results on 1257 pages for 'entity sql'.

Page 818/1257 | < Previous Page | 814 815 816 817 818 819 820 821 822 823 824 825  | Next Page >

  • iphone core data loop array and save each

    - by Matt Facer
    I have a core data model with two tables (meal and ingredients). I am trying to save ONE meal with MANY ingredients. I have the code below which loops through an array of ingredients. I'm trying to save it, but I cannot redeclare the "entity" below. How do I do it? I've tried releasing it, but that didn't work! Thanks for any help. for (x=0;x<ingredients;x++) { NSEntityDescription *entity = [NSEntityDescription insertNewObjectForEntityForName:@"Ingredient" inManagedObjectContext:managedObjectContext]; entity.name = @"test"; } (this method does work saving ONE record out of the loop.. so that's not the problem)

    Read the article

  • Validation errors from Google App Engine Logout link

    - by goggin13
    I am making a web page using the Google App Engine. I am validating my pages, and found that the logout link that is generated by the call to the users api (in python) users.create_logout_url(request.uri) does not validate as XHTML 1.0 Strict. The href in the anchor tag looks like this: /_ah/login?continue=http%3A//localhost%3A8080/&action=Logout Including a link with this anchor text throws three different validation errors: *general entity "action" not defined and no default entity *reference to entity "action" for which no system identifier could be generated *EntityRef: expecting ';' Here is a dummy page with the anchor tag in it, if you want to try it on w3c validator.Dummy Page. The logout link wont work, but you can see how the page is valid without it, but the actual text inside the href tag breaks the validation. Any thoughts on whats going on? Thank you!

    Read the article

  • Echoing a pseudo column value after a COUNT

    - by rob - not a robber
    Hi Gang... Please don't beat me if this is elementary. I searched and found disjointed stuff relating to pseudo columns. Nothing spot on about what I need. Anyway... I have a table with some rows. Each record has a unique ID, an ID that relates to another entity and finally a comment that relates to that last entity. So, I want to COUNT these rows to basically find what entity has the most comments. Instead of me explaining the query, I'll print it SELECT entity_id, COUNT(*) AS amount FROM comments GROUP BY entity_id ORDER BY amount DESC The query does just what I want, but I want to echo the values from that pseudo column, 'amount' Can it be done, or should I use another method like mysql_num_rows? Thank you!!!

    Read the article

  • how to get http get request params in jsf 2.0 bakcing bean?

    - by Marko
    Hi all, I having trouble with passing http get parameters to jsf 2.0 backing bean. User will invoke URl with some params containing id of some entity, which is later used to persist some other entity in db. whole process can be summarized by fallowing: 1. user open page http://www.somhost.com/JsfApp/step-one.xhtml?sid=1 2. user fills some data and goes to next page 3. user fills some more data and then entity is saved to db with sid param from step one. I have session scoped backing bean that hold data from all the pages (steps), but I cant pass param to bean property.. any ideas?

    Read the article

  • MonoRail - Clearing value on edit form submission doesn't trigger validation

    - by Justin
    Hey, In a MonoRail app I have an add/edit view. On add if I don't pick a value for a dropdown I get a validation error and am forced to pick a value. However, if I then edit that same item and reset the dropdown back to the first item ("Select an Item", "0"), it saves and doesn't say it was invalid. Debugging the entity sent back to the server, I can see that it's sending the original value for the dropdown instead of the new value of "0". What's going on here? The first thing I would expect is that it would trigger the validation since the dropdown value isn't set. The second thing I would expect is that it would send the new value I send, not the original. It does send the new value if I change it to another value, but it's as if it has internal logic that says - "I'm updating this entity and a new value was not passed, just don't change the entity value." Thanks, Justin

    Read the article

  • Core data storage is repeated...

    - by Kamlesh
    Hi all, I am trying to use Core Data in my application and I have been succesful in storing data into the entity.The data storage is done in the applicationDidFinishLaunchingWithOptions() method.But when I run the app again,it again gets saved.So how do I check if the data is already present or not?? Here is the code(Saving):-`NSManagedObjectContext *context = [self managedObjectContext]; NSManagedObject *failedBankInfo = [NSEntityDescription insertNewObjectForEntityForName:@"FailedBankInfo" inManagedObjectContext:context]; [failedBankInfo setValue:@"Test Bank" forKey:@"name"]; [failedBankInfo setValue:@"Testville" forKey:@"city"]; [failedBankInfo setValue:@"Testland" forKey:@"state"]; NSError *error; if (![context save:&error]) { NSLog(@"Whoops, couldn't save: %@", [error localizedDescription]); } (Retrieving):- NSFetchRequest *fetchRequest = [[NSFetchRequest alloc] init]; NSEntityDescription *entity = [NSEntityDescription entityForName:@"FailedBankInfo" inManagedObjectContext:context]; [fetchRequest setEntity:entity]; NSArray *fetchedObjects = [context executeFetchRequest:fetchRequest error:&error]; for (NSManagedObject *info in fetchedObjects) { NSLog(@"Name: %@", [info valueForKey:@"name"]); } `

    Read the article

  • Why SELECT N + 1 with no foreign keys and LINQ?

    - by Daniel Flöijer
    I have a database that unfortunately have no real foreign keys (I plan to add this later, but prefer not to do it right now to make migration easier). I have manually written domain objects that map to the database to set up relationships (following this tutorial http://www.codeproject.com/Articles/43025/A-LINQ-Tutorial-Mapping-Tables-to-Objects), and I've finally gotten the code to run properly. However, I've noticed I now have the SELECT N + 1 problem. Instead of selecting all Product's they're selected one by one with this SQL: SELECT [t0].[id] AS [ProductID], [t0].[Name], [t0].[info] AS [Description] FROM [products] AS [t0] WHERE [t0].[id] = @p0 -- @p0: Input Int (Size = -1; Prec = 0; Scale = 0) [65] Controller: public ViewResult List(string category, int page = 1) { var cat = categoriesRepository.Categories.SelectMany(c => c.LocalizedCategories).Where(lc => lc.CountryID == 1).First(lc => lc.Name == category).Category; var productsToShow = cat.Products; var viewModel = new ProductsListViewModel { Products = productsToShow.Skip((page - 1) * PageSize).Take(PageSize).ToList(), PagingInfo = new PagingInfo { CurrentPage = page, ItemsPerPage = PageSize, TotalItems = productsToShow.Count() }, CurrentCategory = cat }; return View("List", viewModel); } Since I wasn't sure if my LINQ expression was correct I tried to just use this but I still got N+1: var cat = categoriesRepository.Categories.First(); Domain objects: [Table(Name = "products")] public class Product { [Column(Name = "id", IsPrimaryKey = true, IsDbGenerated = true, AutoSync = AutoSync.OnInsert)] public int ProductID { get; set; } [Column] public string Name { get; set; } [Column(Name = "info")] public string Description { get; set; } private EntitySet<ProductCategory> _productCategories = new EntitySet<ProductCategory>(); [System.Data.Linq.Mapping.Association(Storage = "_productCategories", OtherKey = "productId", ThisKey = "ProductID")] private ICollection<ProductCategory> ProductCategories { get { return _productCategories; } set { _productCategories.Assign(value); } } public ICollection<Category> Categories { get { return (from pc in ProductCategories select pc.Category).ToList(); } } } [Table(Name = "products_menu")] class ProductCategory { [Column(IsPrimaryKey = true, Name = "products_id")] private int productId; private EntityRef<Product> _product = new EntityRef<Product>(); [System.Data.Linq.Mapping.Association(Storage = "_product", ThisKey = "productId")] public Product Product { get { return _product.Entity; } set { _product.Entity = value; } } [Column(IsPrimaryKey = true, Name = "products_types_id")] private int categoryId; private EntityRef<Category> _category = new EntityRef<Category>(); [System.Data.Linq.Mapping.Association(Storage = "_category", ThisKey = "categoryId")] public Category Category { get { return _category.Entity; } set { _category.Entity = value; } } } [Table(Name = "products_types")] public class Category { [Column(Name = "id", IsPrimaryKey = true, IsDbGenerated = true, AutoSync = AutoSync.OnInsert)] public int CategoryID { get; set; } private EntitySet<ProductCategory> _productCategories = new EntitySet<ProductCategory>(); [System.Data.Linq.Mapping.Association(Storage = "_productCategories", OtherKey = "categoryId", ThisKey = "CategoryID")] private ICollection<ProductCategory> ProductCategories { get { return _productCategories; } set { _productCategories.Assign(value); } } public ICollection<Product> Products { get { return (from pc in ProductCategories select pc.Product).ToList(); } } private EntitySet<LocalizedCategory> _LocalizedCategories = new EntitySet<LocalizedCategory>(); [System.Data.Linq.Mapping.Association(Storage = "_LocalizedCategories", OtherKey = "CategoryID")] public ICollection<LocalizedCategory> LocalizedCategories { get { return _LocalizedCategories; } set { _LocalizedCategories.Assign(value); } } } [Table(Name = "products_types_localized")] public class LocalizedCategory { [Column(Name = "id", IsPrimaryKey = true, IsDbGenerated = true, AutoSync = AutoSync.OnInsert)] public int LocalizedCategoryID { get; set; } [Column(Name = "products_types_id")] private int CategoryID; private EntityRef<Category> _Category = new EntityRef<Category>(); [System.Data.Linq.Mapping.Association(Storage = "_Category", ThisKey = "CategoryID")] public Category Category { get { return _Category.Entity; } set { _Category.Entity = value; } } [Column(Name = "country_id")] public int CountryID { get; set; } [Column] public string Name { get; set; } } I've tried to comment out everything from my View, so nothing there seems to influence this. The ViewModel is as simple as it looks, so shouldn't be anything there. When reading this ( http://www.hookedonlinq.com/LinqToSQL5MinuteOVerview.ashx) I started suspecting it might be because I have no real foreign keys in the database and that I might need to use manual joins in my code. Is that correct? How would I go about it? Should I remove my mapping code from my domain model or is it something that I need to add/change to it? Note: I've stripped parts of the code out that I don't think is relevant to make it cleaner for this question. Please let me know if something is missing.

    Read the article

  • Is it a good idea to close and open hibernate sessions frequently?

    - by Gaurav
    Hi, I'm developing an application which requires that state of entities be read from a database at frequent intervals or triggers. However, once hibernate reads the state, it doesn't re-read it unless I explicitly close the session and read the entity in a new session. Is it a good idea to open a session everytime I want to read the entity and then close it afterwards? How much of an overhead does this put on the application and the database (we use a c3p0 connection pool also)? Will it be enough to simply evict the entity from the session before reading it again?

    Read the article

  • amp is included in url struts tag

    - by lakshmanan
    Hi, In my web application, I use strust2 url tag to pass parameters like id etc., For example, I use a link to delete an entity and I use param to pass the id of the entity to be deleted. And I follow this throughout my web app for adding, editing, deleting an entity. During run time, sometimes, I don't get the params to be stored in my action's bean properties. When I see the link that is generated, I get something like <a href='/projit1/p/discuss/viewDiscussion.action?d=11&amp;amp;amp;projid=11&amp;amp;disid=4'> What are these amps for ? why do they sit in between the action calls (made by link via url tag actions ) ? By the time I traverse back and forth in my web app, I get 10s and 20s of amp sitting in the request URL. What is the problem here ? Please help.

    Read the article

  • AutoMapper determine what to map based on generic type

    - by Daz Lewis
    Hi, Is there a way to provide AutoMapper with just a source and based on the specified mapping for the type of that source automatically determine what to map to? So for example I have a type of Foo and I always want it mapped to Bar but at runtime my code can receive any one of a number of generic types. public T Add(T entity) { //List of mappings var mapList = new Dictionary<Type, Type> { {typeof (Foo), typeof (Bar)} {typeof (Widget), typeof (Sprocket)} }; //Based on the type of T determine what we map to...somehow! var t = mapList[entity.GetType()]; //What goes in ?? to ensure var in the case of Foo will be a Bar? var destination = AutoMapper.Mapper.Map<T, ??>(entity); } Any help is much appreciated.

    Read the article

  • Annotations: methods vs variables

    - by Zenzen
    I was always sure (don't know why) that it's better to add annotations to variables, but while browsing the Hibernate doc http://docs.jboss.org/hibernate/stable/annotations/reference/en/html_single/#entity-hibspec-collection I noticed they tend to annotate the methods. So should I put my annotations before methods, like this: @Entity public class Flight implements Serializable { private long id; @Id @GeneratedValue public long getId() { return id; } public void setId(long id) { this.id = id; } } Or is it better to do it like this: @Entity public class Flight implements Serializable { @Id @GeneratedValue private long id; public long getId() { return id; } public void setId(long id) { this.id = id; } } Or maybe there's no difference?

    Read the article

  • I have to generate PL/SQL using Java. Most of the procedures are common. Only a few keeps changing.

    - by blog
    I have to generate PL-SQL code, with some common code(invariable) and a variable code. I don't want to use any external tools. Some ways that I can think: Can I go and maintain the common code in a template and with markers, where my java code will generate code in the markers and generate a new file. Maintain the common code in static constant String and then generate the whole code in StringBuffer and at last write to file. But, I am not at all satisfied with both the ideas. Can you please suggest any better ways of doing this or the use of any design patterns or anything? Thanks in Advance.

    Read the article

  • Table per subclass inheritance relationship: How to query against the Parent class without loading a

    - by Arthur Ronald F D Garcia
    Suppose a Table per subclass inheritance relationship which can be described bellow (From wikibooks.org - see here) Notice Parent class is not abstract @Entity @Inheritance(strategy=InheritanceType.JOINED) public class Project { @Id private long id; // Other properties } @Entity @Table(name="LARGEPROJECT") public class LargeProject extends Project { private BigDecimal budget; } @Entity @Table(name="SMALLPROJECT") public class SmallProject extends Project { } I have a scenario where i just need to retrieve the Parent class. Because of performance issues, What should i do to run a HQL query in order to retrieve the Parent class and just the Parent class without loading any subclass ???

    Read the article

  • How to programaticly access min and Max values defined in a core-data model designed with XCode ?

    - by Xav
    I was expecting to find that in the NSAttributeDescription class, but only the default value is there. Behind the scene I tought a validationPredicate was created but trying to reach it using NSDictionary* dico= [[myManagedObject entity] propertiesByName]; NSAttributeDescription* attributeDescription=[dico objectForKey:attributeKey]; for (NSString* string in [attributeDescription validationWarnings]) just get me nowhere, no validationWarnings, no validationPredicates... any thoughts on this ? Edit1: It seems that getting the entity straight from the managedObject doesn't give you the full picture. Getting the Entity from the NSManagedObjectModel permits to reach the validationWarnings & validationPredicates...

    Read the article

  • Many-to-one relation exception due to closed session after loading

    - by Nick Thissen
    Hi, I am using NHibernate (version 1.2.1) for the first time so I wrote a simple test application (an ASP.NET project) that uses it. In my database I have two tables: Persons and Categories. Each person gets one category, seems easy enough. | Persons | | Categories | |--------------| |--------------| | Id (PK) | | Id (PK) | | Firstname | | CategoryName | | Lastname | | CreatedTime | | CategoryId | | UpdatedTime | | CreatedTime | | Deleted | | UpdatedTime | | Deleted | The Id, CreatedTime, UpdatedTime and Deleted attributes are a convention I use in all my tables, so I have tried to bring this fact into an additional abstraction layer. I have a project DatabaseFramework which has three important classes: Entity: an abstract class that defines these four properties. All 'entity objects' (in this case Person and Category) must inherit Entity. IEntityManager: a generic interface (type parameter as Entity) that defines methods like Load, Insert, Update, etc. NHibernateEntityManager: an implementation of this interface using NHibernate to do the loading, saving, etc. Now, the Person and Category classes are straightforward, they just define the attributes of the tables of course (keeping in mind that four of them are in the base Entity class). Since the Persons table is related to the Categories table via the CategoryId attribute, the Person class has a Category property that holds the related category. However, in my webpage, I will also need the name of this category (CategoryName), for databinding purposes for example. So I created an additional property CategoryName that returns the CategoryName property of the current Category property, or an empty string if the Category is null: Namespace Database Public Class Person Inherits DatabaseFramework.Entity Public Overridable Property Firstname As String Public Overridable Property Lastname As String Public Overridable Property Category As Category Public Overridable ReadOnly Property CategoryName As String Get Return If(Me.Category Is Nothing, _ String.Empty, _ Me.Category.CategoryName) End Get End Property End Class End Namespace I am mapping the Person class using this mapping file. The many-to-one relation was suggested by Yads in another thread: <id name="Id" column="Id" type="int" unsaved-value="0"> <generator class="identity" /> </id> <property name="CreatedTime" type="DateTime" not-null="true" /> <property name="UpdatedTime" type="DateTime" not-null="true" /> <property name="Deleted" type="Boolean" not-null="true" /> <property name="Firstname" type="String" /> <property name="Lastname" type="String" /> <many-to-one name="Category" column="CategoryId" class="NHibernateWebTest.Database.Category, NHibernateWebTest" /> (I can't get it to show the root node, this forum hides it, I don't know how to escape the html-like tags...) The final important detail is the Load method of the NHibernateEntityManager implementation. (This is in C# as it's in a different project, sorry about that). I simply open a new ISession (ISessionFactory.OpenSession) in the GetSession method and then use that to fill an EntityCollection(Of TEntity) which is just a collection inheriting System.Collections.ObjectModel.Collection(Of T). public virtual EntityCollection< TEntity Load() { using (ISession session = this.GetSession()) { var entities = session .CreateCriteria(typeof (TEntity)) .Add(Expression.Eq("Deleted", false)) .List< TEntity (); return new EntityCollection< TEntity (entities); } } (Again, I can't get it to format the code correctly, it hides the generic type parameters, probably because it reads the angled symbols as a HTML tag..? If you know how to let me do that, let me know!) Now, the idea of this Load method is that I get a fully functional collection of Persons, all their properties set to the correct values (including the Category property, and thus, the CategoryName property should return the correct name). However, it seems that is not the case. When I try to data-bind the result of this Load method to a GridView in ASP.NET, it tells me this: Property accessor 'CategoryName' on object 'NHibernateWebTest.Database.Person' threw the following exception:'Could not initialize proxy - the owning Session was closed.' The exception occurs on the DataBind method call here: public virtual void LoadGrid() { if (this.Grid == null) return; this.Grid.DataSource = this.Manager.Load(); this.Grid.DataBind(); } Well, of course the session is closed, I closed it via the using block. Isn't that the correct approach, should I keep the session open? And for how long? Can I close it after the DataBind method has been run? In each case, I'd really like my Load method to just return a functional collection of items. It seems to me that it is now only getting the Category when it is required (eg, when the GridView wants to read the CategoryName, which wants to read the Category property), but at that time the session is closed. Is that reasoning correct? How do I stop this behavior? Or shouldn't I? And what should I do otherwise? Thanks!

    Read the article

  • How can one change the primary key using NHibernate.

    - by mark
    Dear ladies and sirs. I want to change the primary key of an entity in database, so that all the relevant foreign key constraints are updated as well./We are using NHibenate as our ORM. Is it possible to do it? Thanks. P.S. I know the practice of changing the primary key is highly discouraged. My problem is that my primary key is backed by a natural Id of the entity, which may sometimes change. We could, theoretically, utilize a unique primary key, unrelated to the natural key of the entity, but this complicates things too much in other places, so this is not an option.

    Read the article

  • afterTransactionCompletion not working

    - by Attilah
    I created an hibernate interceptor : public class MyInterceptor extends EmptyInterceptor { private boolean isCanal=false; public boolean onSave(Object entity, Serializable arg1, Object[] arg2, String[] arg3, Type[] arg4) throws CallbackException { for(int i=0;i<100;i++){ System.out.println("Inside MyInterceptor(onSave) : "+entity.toString()); } if(entity instanceof Canal){ isCanal=true; } return false; } public void afterTransactionCompletion(Transaction tx){ if(tx.wasCommitted()&&(isCanal)){ for(int i=0;i<100;i++){ System.out.println("Inside MyInterceptor(afterTransactionCompletion) : Canal was saved to DB."); } } } but the method afterTransactionCompletion doesn't get executed after a transaction is commited. I've tried all the ways I know of but I can't make it work. What's more surprising is that the onSave method works fine. Help ! Could this be due to this bug ? : http://opensource.atlassian.com/projects/hibernate/browse/HHH-1956 How can I circumvent this bug if it's the cause ?

    Read the article

  • Why is Hibernate not loading a column?

    - by B.R.
    I've got an entity with a few properties that gets used a lot in my Hibernate/GWT app. For the most part, everything works fine, but Hibernate refuses to load one of the properties. It doesn't appear in the query, despite being annotated correctly in the entity. The relevant portion of the entity: @Column(name="HasSubSlots") @Type(type="yes_no") public boolean hasSubSlotSupport() { return hasSubSlotSupport; } And the generated SQL query: Hibernate: /* load entities.DeviceModel */ select devicemode0_.DevModel as DevModel1_0_, devicemode0_.InvModelName as InvModel2_1_0_ from DeviceModels devicemode0_ where devicemode0_.DevModel=? Despite the fact that I refer to that property, it's never loaded, lazily or not, and the getter always returns false. Any ideas on how I can dig deeper into this, or what might be wrong?

    Read the article

  • DocumentBuilder.parse() / Parsing Entities

    - by stormin986
    I'm new to parsing XML and am having an issue with entities. (Am doing this on Android, if it makes a difference). Is there a way to have it turn an entity into the character it represents? I have this in the child of an element: "isn&#39;t" (minus quotes). I would prefer it parse it and the end result be a single text node. However, right now this is turned in to TEXT, ENTITY, TEXT. Is there a way to automatically have it parse the entity into text, or a manual way to do it?

    Read the article

  • Oracle Hibernate with in Netbean RCP

    - by jurnaltejo
    All, i have a problem with hibernate using netbean platform 6.8, i have been search around internet, but cannot found the suitable answer This is my story. i am using oracle database as data source of my hibernate entity with ojdbc14.jar driver. First i create hibernate entity tobe wrapped latter in a netbeans module, i tested the hibernate connection configuration and everything just works well. i can connect to oracle database successfuly, every hibernate query works well. Then i wrapped that hibernate entity jar as a netbeans module, create another module to warp my ojdbc14.jar then i test it. and, im using hibernate library dependency that available on netbean platform (netbean 6.8), but unfutornatelly i got oracle sql error saying “no suitable driver for [connection url]” when running the project. thats quite weird since it doesn’t happend when I test it before with out netbean platform. i thought that is related to netbeans lazy loading issue, i am not sure,. any idea ? tq for help

    Read the article

  • OneToMany association updates instead of insert

    - by Shvalb
    I have an entity with one-to-many association to child entity. The child entity has 2 columns as PK and one of the column is FK to the parent table. mapping looks like this: @OneToMany(cascade = {CascadeType.ALL}, fetch = FetchType.EAGER ) @JoinColumn(name="USER_RESULT_SEQUENCES.USER_RESULT_ID", referencedColumnName="USER_RESULT_ID", unique=true, insertable=true, updatable=false) private List<UserResultSequence> sequences; I create an instance of parent and add children instances to list and then try to save it to DB. If child table is empty it inserts all children and it works perfectly. if the child table is not empty it updates existing rows! I don't know why it updates instead of inserts, any ideas why this might happen?? Thank you!

    Read the article

  • Parsing SQLIO Output to Excel Charts using Regex in PowerShell

    - by Jonathan Kehayias
    Today Joe Webb ( Blog | Twitter ) blogged about The Power of Regex in Powershell, and in his post he shows how to parse the SQL Server Error Log for events of interest. At the end of his blog post Joe asked about other places where Regular Expressions have been useful in PowerShell so I thought I’d blog my script for parsing SQLIO output using Regex in PowerShell, to populate an Excel worksheet and build charts based on the results automatically. If you’ve never used SQLIO, Brent Ozar ( Blog | Twitter...(read more)

    Read the article

  • Parsing SQLIO Output to Excel Charts using Regex in PowerShell

    - by Jonathan Kehayias
    Today Joe Webb ( Blog | Twitter ) blogged about The Power of Regex in Powershell, and in his post he shows how to parse the SQL Server Error Log for events of interest.  At the end of his blog post Joe asked about other places where Regular Expressions have been useful in PowerShell so I thought I’d blog my script for parsing SQLIO output using Regex in PowerShell, to populate an Excel worksheet and build charts based on the results automatically. If you’ve never used SQLIO, Brent Ozar ( Blog...(read more)

    Read the article

  • Reporting Services - It's a Wrap!

    - by smisner
    If you have any experience at all with Reporting Services, you have probably developed a report using the matrix data region. It's handy when you want to generate columns dynamically based on data. If users view a matrix report online, they can scroll horizontally to view all columns and all is well. But if they want to print the report, the experience is completely different and you'll have to decide how you want to handle dynamic columns. By default, when a user prints a matrix report for which the number of columns exceeds the width of the page, Reporting Services determines how many columns can fit on the page and renders one or more separate pages for the additional columns. In this post, I'll explain two techniques for managing dynamic columns. First, I'll show how to use the RepeatRowHeaders property to make it easier to read a report when columns span multiple pages, and then I'll show you how to "wrap" columns so that you can avoid the horizontal page break. Included with this post are the sample RDLs for download. First, let's look at the default behavior of a matrix. A matrix that has too many columns for one printed page (or output to page-based renderer like PDF or Word) will be rendered such that the first page with the row group headers and the inital set of columns, as shown in Figure 1. The second page continues by rendering the next set of columns that can fit on the page, as shown in Figure 2.This pattern continues until all columns are rendered. The problem with the default behavior is that you've lost the context of employee and sales order - the row headers - on the second page. That makes it hard for users to read this report because the layout requires them to flip back and forth between the current page and the first page of the report. You can fix this behavior by finding the RepeatRowHeaders of the tablix report item and changing its value to True. The second (and subsequent pages) of the matrix now look like the image shown in Figure 3. The problem with this approach is that the number of printed pages to flip through is unpredictable when you have a large number of potential columns. What if you want to include all columns on the same page? You can take advantage of the repeating behavior of a tablix and get repeating columns by embedding one tablix inside of another. For this example, I'm using SQL Server 2008 R2 Reporting Services. You can get similar results with SQL Server 2008. (In fact, you could probably do something similar in SQL Server 2005, but I haven't tested it. The steps would be slightly different because you would be working with the old-style matrix as compared to the new-style tablix discussed in this post.) I created a dataset that queries AdventureWorksDW2008 tables: SELECT TOP (100) e.LastName + ', ' + e.FirstName AS EmployeeName, d.FullDateAlternateKey, f.SalesOrderNumber, p.EnglishProductName, sum(SalesAmount) as SalesAmount FROM FactResellerSales AS f INNER JOIN DimProduct AS p ON p.ProductKey = f.ProductKey INNER JOIN DimDate AS d ON d.DateKey = f.OrderDateKey INNER JOIN DimEmployee AS e ON e.EmployeeKey = f.EmployeeKey GROUP BY p.EnglishProductName, d.FullDateAlternateKey, e.LastName + ', ' + e.FirstName, f.SalesOrderNumber ORDER BY EmployeeName, f.SalesOrderNumber, p.EnglishProductName To start the report: Add a matrix to the report body and drag Employee Name to the row header, which also creates a group. Next drag SalesOrderNumber below Employee Name in the Row Groups panel, which creates a second group and a second column in the row header section of the matrix, as shown in Figure 4. Now for some trickiness. Add another column to the row headers. This new column will be associated with the existing EmployeeName group rather than causing BIDS to create a new group. To do this, right-click on the EmployeeName textbox in the bottom row, point to Insert Column, and then click Inside Group-Right. Then add the SalesOrderNumber field to this new column. By doing this, you're creating a report that repeats a set of columns for each EmployeeName/SalesOrderNumber combination that appears in the data. Next, modify the first row group's expression to group on both EmployeeName and SalesOrderNumber. In the Row Groups section, right-click EmployeeName, click Group Properties, click the Add button, and select [SalesOrderNumber]. Now you need to configure the columns to repeat. Rather than use the Columns group of the matrix like you might expect, you're going to use the textbox that belongs to the second group of the tablix as a location for embedding other report items. First, clear out the text that's currently in the third column - SalesOrderNumber - because it's already added as a separate textbox in this report design. Then drag and drop a matrix into that textbox, as shown in Figure 5. Again, you need to do some tricks here to get the appearance and behavior right. We don't really want repeating rows in the embedded matrix, so follow these steps: Click on the Rows label which then displays RowGroup in the Row Groups pane below the report body. Right-click on RowGroup,click Delete Group, and select the option to delete associated rows and columns. As a result, you get a modified matrix which has only a ColumnGroup in it, with a row above a double-dashed line for the column group and a row below the line for the aggregated data. Let's continue: Drag EnglishProductName to the data textbox (below the line). Add a second data row by right-clicking EnglishProductName, pointing to Insert Row, and clicking Below. Add the SalesAmount field to the new data textbox. Now eliminate the column group row without eliminating the group. To do this, right-click the row above the double-dashed line, click Delete Rows, and then select Delete Rows Only in the message box. Now you're ready for the fit and finish phase: Resize the column containing the embedded matrix so that it fits completely. Also, the final column in the matrix is for the column group. You can't delete this column, but you can make it as small as possible. Just click on the matrix to display the row and column handles, and then drag the right edge of the rightmost column to the left to make the column virtually disappear. Next, configure the groups so that the columns of the embedded matrix will wrap. In the Column Groups pane, right-click ColumnGroup1 and click on the expression button (labeled fx) to the right of Group On [EnglishProductName]. Replace the expression with the following: =RowNumber("SalesOrderNumber" ). We use SalesOrderNumber here because that is the name of the group that "contains" the embedded matrix. The next step is to configure the number of columns to display before wrapping. Click any cell in the matrix that is not inside the embedded matrix, and then double-click the second group in the Row Groups pane - SalesOrderNumber. Change the group expression to the following expression: =Ceiling(RowNumber("EmployeeName")/3) The last step is to apply formatting. In my example, I set the SalesAmount textbox's Format property to C2 and also right-aligned the text in both the EnglishProductName and the SalesAmount textboxes. And voila - Figure 6 shows a matrix report with wrapping columns. Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • Improving Partitioned Table Join Performance

    - by Paul White
    The query optimizer does not always choose an optimal strategy when joining partitioned tables. This post looks at an example, showing how a manual rewrite of the query can almost double performance, while reducing the memory grant to almost nothing. Test Data The two tables in this example use a common partitioning partition scheme. The partition function uses 41 equal-size partitions: CREATE PARTITION FUNCTION PFT (integer) AS RANGE RIGHT FOR VALUES ( 125000, 250000, 375000, 500000, 625000, 750000, 875000, 1000000, 1125000, 1250000, 1375000, 1500000, 1625000, 1750000, 1875000, 2000000, 2125000, 2250000, 2375000, 2500000, 2625000, 2750000, 2875000, 3000000, 3125000, 3250000, 3375000, 3500000, 3625000, 3750000, 3875000, 4000000, 4125000, 4250000, 4375000, 4500000, 4625000, 4750000, 4875000, 5000000 ); GO CREATE PARTITION SCHEME PST AS PARTITION PFT ALL TO ([PRIMARY]); There two tables are: CREATE TABLE dbo.T1 ( TID integer NOT NULL IDENTITY(0,1), Column1 integer NOT NULL, Padding binary(100) NOT NULL DEFAULT 0x,   CONSTRAINT PK_T1 PRIMARY KEY CLUSTERED (TID) ON PST (TID) );   CREATE TABLE dbo.T2 ( TID integer NOT NULL, Column1 integer NOT NULL, Padding binary(100) NOT NULL DEFAULT 0x,   CONSTRAINT PK_T2 PRIMARY KEY CLUSTERED (TID, Column1) ON PST (TID) ); The next script loads 5 million rows into T1 with a pseudo-random value between 1 and 5 for Column1. The table is partitioned on the IDENTITY column TID: INSERT dbo.T1 WITH (TABLOCKX) (Column1) SELECT (ABS(CHECKSUM(NEWID())) % 5) + 1 FROM dbo.Numbers AS N WHERE n BETWEEN 1 AND 5000000; In case you don’t already have an auxiliary table of numbers lying around, here’s a script to create one with 10 million rows: CREATE TABLE dbo.Numbers (n bigint PRIMARY KEY);   WITH L0 AS(SELECT 1 AS c UNION ALL SELECT 1), L1 AS(SELECT 1 AS c FROM L0 AS A CROSS JOIN L0 AS B), L2 AS(SELECT 1 AS c FROM L1 AS A CROSS JOIN L1 AS B), L3 AS(SELECT 1 AS c FROM L2 AS A CROSS JOIN L2 AS B), L4 AS(SELECT 1 AS c FROM L3 AS A CROSS JOIN L3 AS B), L5 AS(SELECT 1 AS c FROM L4 AS A CROSS JOIN L4 AS B), Nums AS(SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) AS n FROM L5) INSERT dbo.Numbers WITH (TABLOCKX) SELECT TOP (10000000) n FROM Nums ORDER BY n OPTION (MAXDOP 1); Table T1 contains data like this: Next we load data into table T2. The relationship between the two tables is that table 2 contains ‘n’ rows for each row in table 1, where ‘n’ is determined by the value in Column1 of table T1. There is nothing particularly special about the data or distribution, by the way. INSERT dbo.T2 WITH (TABLOCKX) (TID, Column1) SELECT T.TID, N.n FROM dbo.T1 AS T JOIN dbo.Numbers AS N ON N.n >= 1 AND N.n <= T.Column1; Table T2 ends up containing about 15 million rows: The primary key for table T2 is a combination of TID and Column1. The data is partitioned according to the value in column TID alone. Partition Distribution The following query shows the number of rows in each partition of table T1: SELECT PartitionID = CA1.P, NumRows = COUNT_BIG(*) FROM dbo.T1 AS T CROSS APPLY (VALUES ($PARTITION.PFT(TID))) AS CA1 (P) GROUP BY CA1.P ORDER BY CA1.P; There are 40 partitions containing 125,000 rows (40 * 125k = 5m rows). The rightmost partition remains empty. The next query shows the distribution for table 2: SELECT PartitionID = CA1.P, NumRows = COUNT_BIG(*) FROM dbo.T2 AS T CROSS APPLY (VALUES ($PARTITION.PFT(TID))) AS CA1 (P) GROUP BY CA1.P ORDER BY CA1.P; There are roughly 375,000 rows in each partition (the rightmost partition is also empty): Ok, that’s the test data done. Test Query and Execution Plan The task is to count the rows resulting from joining tables 1 and 2 on the TID column: SET STATISTICS IO ON; DECLARE @s datetime2 = SYSUTCDATETIME();   SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID;   SELECT DATEDIFF(Millisecond, @s, SYSUTCDATETIME()); SET STATISTICS IO OFF; The optimizer chooses a plan using parallel hash join, and partial aggregation: The Plan Explorer plan tree view shows accurate cardinality estimates and an even distribution of rows across threads (click to enlarge the image): With a warm data cache, the STATISTICS IO output shows that no physical I/O was needed, and all 41 partitions were touched: Running the query without actual execution plan or STATISTICS IO information for maximum performance, the query returns in around 2600ms. Execution Plan Analysis The first step toward improving on the execution plan produced by the query optimizer is to understand how it works, at least in outline. The two parallel Clustered Index Scans use multiple threads to read rows from tables T1 and T2. Parallel scan uses a demand-based scheme where threads are given page(s) to scan from the table as needed. This arrangement has certain important advantages, but does result in an unpredictable distribution of rows amongst threads. The point is that multiple threads cooperate to scan the whole table, but it is impossible to predict which rows end up on which threads. For correct results from the parallel hash join, the execution plan has to ensure that rows from T1 and T2 that might join are processed on the same thread. For example, if a row from T1 with join key value ‘1234’ is placed in thread 5’s hash table, the execution plan must guarantee that any rows from T2 that also have join key value ‘1234’ probe thread 5’s hash table for matches. The way this guarantee is enforced in this parallel hash join plan is by repartitioning rows to threads after each parallel scan. The two repartitioning exchanges route rows to threads using a hash function over the hash join keys. The two repartitioning exchanges use the same hash function so rows from T1 and T2 with the same join key must end up on the same hash join thread. Expensive Exchanges This business of repartitioning rows between threads can be very expensive, especially if a large number of rows is involved. The execution plan selected by the optimizer moves 5 million rows through one repartitioning exchange and around 15 million across the other. As a first step toward removing these exchanges, consider the execution plan selected by the optimizer if we join just one partition from each table, disallowing parallelism: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = 1 AND $PARTITION.PFT(T2.TID) = 1 OPTION (MAXDOP 1); The optimizer has chosen a (one-to-many) merge join instead of a hash join. The single-partition query completes in around 100ms. If everything scaled linearly, we would expect that extending this strategy to all 40 populated partitions would result in an execution time around 4000ms. Using parallelism could reduce that further, perhaps to be competitive with the parallel hash join chosen by the optimizer. This raises a question. If the most efficient way to join one partition from each of the tables is to use a merge join, why does the optimizer not choose a merge join for the full query? Forcing a Merge Join Let’s force the optimizer to use a merge join on the test query using a hint: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (MERGE JOIN); This is the execution plan selected by the optimizer: This plan results in the same number of logical reads reported previously, but instead of 2600ms the query takes 5000ms. The natural explanation for this drop in performance is that the merge join plan is only using a single thread, whereas the parallel hash join plan could use multiple threads. Parallel Merge Join We can get a parallel merge join plan using the same query hint as before, and adding trace flag 8649: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (MERGE JOIN, QUERYTRACEON 8649); The execution plan is: This looks promising. It uses a similar strategy to distribute work across threads as seen for the parallel hash join. In practice though, performance is disappointing. On a typical run, the parallel merge plan runs for around 8400ms; slower than the single-threaded merge join plan (5000ms) and much worse than the 2600ms for the parallel hash join. We seem to be going backwards! The logical reads for the parallel merge are still exactly the same as before, with no physical IOs. The cardinality estimates and thread distribution are also still very good (click to enlarge): A big clue to the reason for the poor performance is shown in the wait statistics (captured by Plan Explorer Pro): CXPACKET waits require careful interpretation, and are most often benign, but in this case excessive waiting occurs at the repartitioning exchanges. Unlike the parallel hash join, the repartitioning exchanges in this plan are order-preserving ‘merging’ exchanges (because merge join requires ordered inputs): Parallelism works best when threads can just grab any available unit of work and get on with processing it. Preserving order introduces inter-thread dependencies that can easily lead to significant waits occurring. In extreme cases, these dependencies can result in an intra-query deadlock, though the details of that will have to wait for another time to explore in detail. The potential for waits and deadlocks leads the query optimizer to cost parallel merge join relatively highly, especially as the degree of parallelism (DOP) increases. This high costing resulted in the optimizer choosing a serial merge join rather than parallel in this case. The test results certainly confirm its reasoning. Collocated Joins In SQL Server 2008 and later, the optimizer has another available strategy when joining tables that share a common partition scheme. This strategy is a collocated join, also known as as a per-partition join. It can be applied in both serial and parallel execution plans, though it is limited to 2-way joins in the current optimizer. Whether the optimizer chooses a collocated join or not depends on cost estimation. The primary benefits of a collocated join are that it eliminates an exchange and requires less memory, as we will see next. Costing and Plan Selection The query optimizer did consider a collocated join for our original query, but it was rejected on cost grounds. The parallel hash join with repartitioning exchanges appeared to be a cheaper option. There is no query hint to force a collocated join, so we have to mess with the costing framework to produce one for our test query. Pretending that IOs cost 50 times more than usual is enough to convince the optimizer to use collocated join with our test query: -- Pretend IOs are 50x cost temporarily DBCC SETIOWEIGHT(50);   -- Co-located hash join SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (RECOMPILE);   -- Reset IO costing DBCC SETIOWEIGHT(1); Collocated Join Plan The estimated execution plan for the collocated join is: The Constant Scan contains one row for each partition of the shared partitioning scheme, from 1 to 41. The hash repartitioning exchanges seen previously are replaced by a single Distribute Streams exchange using Demand partitioning. Demand partitioning means that the next partition id is given to the next parallel thread that asks for one. My test machine has eight logical processors, and all are available for SQL Server to use. As a result, there are eight threads in the single parallel branch in this plan, each processing one partition from each table at a time. Once a thread finishes processing a partition, it grabs a new partition number from the Distribute Streams exchange…and so on until all partitions have been processed. It is important to understand that the parallel scans in this plan are different from the parallel hash join plan. Although the scans have the same parallelism icon, tables T1 and T2 are not being co-operatively scanned by multiple threads in the same way. Each thread reads a single partition of T1 and performs a hash match join with the same partition from table T2. The properties of the two Clustered Index Scans show a Seek Predicate (unusual for a scan!) limiting the rows to a single partition: The crucial point is that the join between T1 and T2 is on TID, and TID is the partitioning column for both tables. A thread that processes partition ‘n’ is guaranteed to see all rows that can possibly join on TID for that partition. In addition, no other thread will see rows from that partition, so this removes the need for repartitioning exchanges. CPU and Memory Efficiency Improvements The collocated join has removed two expensive repartitioning exchanges and added a single exchange processing 41 rows (one for each partition id). Remember, the parallel hash join plan exchanges had to process 5 million and 15 million rows. The amount of processor time spent on exchanges will be much lower in the collocated join plan. In addition, the collocated join plan has a maximum of 8 threads processing single partitions at any one time. The 41 partitions will all be processed eventually, but a new partition is not started until a thread asks for it. Threads can reuse hash table memory for the new partition. The parallel hash join plan also had 8 hash tables, but with all 5,000,000 build rows loaded at the same time. The collocated plan needs memory for only 8 * 125,000 = 1,000,000 rows at any one time. Collocated Hash Join Performance The collated join plan has disappointing performance in this case. The query runs for around 25,300ms despite the same IO statistics as usual. This is much the worst result so far, so what went wrong? It turns out that cardinality estimation for the single partition scans of table T1 is slightly low. The properties of the Clustered Index Scan of T1 (graphic immediately above) show the estimation was for 121,951 rows. This is a small shortfall compared with the 125,000 rows actually encountered, but it was enough to cause the hash join to spill to physical tempdb: A level 1 spill doesn’t sound too bad, until you realize that the spill to tempdb probably occurs for each of the 41 partitions. As a side note, the cardinality estimation error is a little surprising because the system tables accurately show there are 125,000 rows in every partition of T1. Unfortunately, the optimizer uses regular column and index statistics to derive cardinality estimates here rather than system table information (e.g. sys.partitions). Collocated Merge Join We will never know how well the collocated parallel hash join plan might have worked without the cardinality estimation error (and the resulting 41 spills to tempdb) but we do know: Merge join does not require a memory grant; and Merge join was the optimizer’s preferred join option for a single partition join Putting this all together, what we would really like to see is the same collocated join strategy, but using merge join instead of hash join. Unfortunately, the current query optimizer cannot produce a collocated merge join; it only knows how to do collocated hash join. So where does this leave us? CROSS APPLY sys.partitions We can try to write our own collocated join query. We can use sys.partitions to find the partition numbers, and CROSS APPLY to get a count per partition, with a final step to sum the partial counts. The following query implements this idea: SELECT row_count = SUM(Subtotals.cnt) FROM ( -- Partition numbers SELECT p.partition_number FROM sys.partitions AS p WHERE p.[object_id] = OBJECT_ID(N'T1', N'U') AND p.index_id = 1 ) AS P CROSS APPLY ( -- Count per collocated join SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals; The estimated plan is: The cardinality estimates aren’t all that good here, especially the estimate for the scan of the system table underlying the sys.partitions view. Nevertheless, the plan shape is heading toward where we would like to be. Each partition number from the system table results in a per-partition scan of T1 and T2, a one-to-many Merge Join, and a Stream Aggregate to compute the partial counts. The final Stream Aggregate just sums the partial counts. Execution time for this query is around 3,500ms, with the same IO statistics as always. This compares favourably with 5,000ms for the serial plan produced by the optimizer with the OPTION (MERGE JOIN) hint. This is another case of the sum of the parts being less than the whole – summing 41 partial counts from 41 single-partition merge joins is faster than a single merge join and count over all partitions. Even so, this single-threaded collocated merge join is not as quick as the original parallel hash join plan, which executed in 2,600ms. On the positive side, our collocated merge join uses only one logical processor and requires no memory grant. The parallel hash join plan used 16 threads and reserved 569 MB of memory:   Using a Temporary Table Our collocated merge join plan should benefit from parallelism. The reason parallelism is not being used is that the query references a system table. We can work around that by writing the partition numbers to a temporary table (or table variable): SET STATISTICS IO ON; DECLARE @s datetime2 = SYSUTCDATETIME();   CREATE TABLE #P ( partition_number integer PRIMARY KEY);   INSERT #P (partition_number) SELECT p.partition_number FROM sys.partitions AS p WHERE p.[object_id] = OBJECT_ID(N'T1', N'U') AND p.index_id = 1;   SELECT row_count = SUM(Subtotals.cnt) FROM #P AS p CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals;   DROP TABLE #P;   SELECT DATEDIFF(Millisecond, @s, SYSUTCDATETIME()); SET STATISTICS IO OFF; Using the temporary table adds a few logical reads, but the overall execution time is still around 3500ms, indistinguishable from the same query without the temporary table. The problem is that the query optimizer still doesn’t choose a parallel plan for this query, though the removal of the system table reference means that it could if it chose to: In fact the optimizer did enter the parallel plan phase of query optimization (running search 1 for a second time): Unfortunately, the parallel plan found seemed to be more expensive than the serial plan. This is a crazy result, caused by the optimizer’s cost model not reducing operator CPU costs on the inner side of a nested loops join. Don’t get me started on that, we’ll be here all night. In this plan, everything expensive happens on the inner side of a nested loops join. Without a CPU cost reduction to compensate for the added cost of exchange operators, candidate parallel plans always look more expensive to the optimizer than the equivalent serial plan. Parallel Collocated Merge Join We can produce the desired parallel plan using trace flag 8649 again: SELECT row_count = SUM(Subtotals.cnt) FROM #P AS p CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals OPTION (QUERYTRACEON 8649); The actual execution plan is: One difference between this plan and the collocated hash join plan is that a Repartition Streams exchange operator is used instead of Distribute Streams. The effect is similar, though not quite identical. The Repartition uses round-robin partitioning, meaning the next partition id is pushed to the next thread in sequence. The Distribute Streams exchange seen earlier used Demand partitioning, meaning the next partition id is pulled across the exchange by the next thread that is ready for more work. There are subtle performance implications for each partitioning option, but going into that would again take us too far off the main point of this post. Performance The important thing is the performance of this parallel collocated merge join – just 1350ms on a typical run. The list below shows all the alternatives from this post (all timings include creation, population, and deletion of the temporary table where appropriate) from quickest to slowest: Collocated parallel merge join: 1350ms Parallel hash join: 2600ms Collocated serial merge join: 3500ms Serial merge join: 5000ms Parallel merge join: 8400ms Collated parallel hash join: 25,300ms (hash spill per partition) The parallel collocated merge join requires no memory grant (aside from a paltry 1.2MB used for exchange buffers). This plan uses 16 threads at DOP 8; but 8 of those are (rather pointlessly) allocated to the parallel scan of the temporary table. These are minor concerns, but it turns out there is a way to address them if it bothers you. Parallel Collocated Merge Join with Demand Partitioning This final tweak replaces the temporary table with a hard-coded list of partition ids (dynamic SQL could be used to generate this query from sys.partitions): SELECT row_count = SUM(Subtotals.cnt) FROM ( VALUES (1),(2),(3),(4),(5),(6),(7),(8),(9),(10), (11),(12),(13),(14),(15),(16),(17),(18),(19),(20), (21),(22),(23),(24),(25),(26),(27),(28),(29),(30), (31),(32),(33),(34),(35),(36),(37),(38),(39),(40),(41) ) AS P (partition_number) CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals OPTION (QUERYTRACEON 8649); The actual execution plan is: The parallel collocated hash join plan is reproduced below for comparison: The manual rewrite has another advantage that has not been mentioned so far: the partial counts (per partition) can be computed earlier than the partial counts (per thread) in the optimizer’s collocated join plan. The earlier aggregation is performed by the extra Stream Aggregate under the nested loops join. The performance of the parallel collocated merge join is unchanged at around 1350ms. Final Words It is a shame that the current query optimizer does not consider a collocated merge join (Connect item closed as Won’t Fix). The example used in this post showed an improvement in execution time from 2600ms to 1350ms using a modestly-sized data set and limited parallelism. In addition, the memory requirement for the query was almost completely eliminated  – down from 569MB to 1.2MB. The problem with the parallel hash join selected by the optimizer is that it attempts to process the full data set all at once (albeit using eight threads). It requires a large memory grant to hold all 5 million rows from table T1 across the eight hash tables, and does not take advantage of the divide-and-conquer opportunity offered by the common partitioning. The great thing about the collocated join strategies is that each parallel thread works on a single partition from both tables, reading rows, performing the join, and computing a per-partition subtotal, before moving on to a new partition. From a thread’s point of view… If you have trouble visualizing what is happening from just looking at the parallel collocated merge join execution plan, let’s look at it again, but from the point of view of just one thread operating between the two Parallelism (exchange) operators. Our thread picks up a single partition id from the Distribute Streams exchange, and starts a merge join using ordered rows from partition 1 of table T1 and partition 1 of table T2. By definition, this is all happening on a single thread. As rows join, they are added to a (per-partition) count in the Stream Aggregate immediately above the Merge Join. Eventually, either T1 (partition 1) or T2 (partition 1) runs out of rows and the merge join stops. The per-partition count from the aggregate passes on through the Nested Loops join to another Stream Aggregate, which is maintaining a per-thread subtotal. Our same thread now picks up a new partition id from the exchange (say it gets id 9 this time). The count in the per-partition aggregate is reset to zero, and the processing of partition 9 of both tables proceeds just as it did for partition 1, and on the same thread. Each thread picks up a single partition id and processes all the data for that partition, completely independently from other threads working on other partitions. One thread might eventually process partitions (1, 9, 17, 25, 33, 41) while another is concurrently processing partitions (2, 10, 18, 26, 34) and so on for the other six threads at DOP 8. The point is that all 8 threads can execute independently and concurrently, continuing to process new partitions until the wider job (of which the thread has no knowledge!) is done. This divide-and-conquer technique can be much more efficient than simply splitting the entire workload across eight threads all at once. Related Reading Understanding and Using Parallelism in SQL Server Parallel Execution Plans Suck © 2013 Paul White – All Rights Reserved Twitter: @SQL_Kiwi

    Read the article

< Previous Page | 814 815 816 817 818 819 820 821 822 823 824 825  | Next Page >