SQL.RU
 client/server technologies
 Главная | Документация | Статьи | Книги | Форум | Блоги | Опросы | Гостевая | Рассылка | Работа | Поиск | FAQ |
   
#346<<  #347  >>#348

Дни разработчика - весна 2007

СОДЕРЖАНИЕ

1.СЕМИНАР
1.1.Microsoft Business Intelligence в деталях для ИТ - профессионалов
2.СТАТЬИ
2.1.Основы ввода-вывода Microsoft SQL Server. Часть 2 (продолжение)
3.ССЫЛКИ НА СТАТЬИ
3.1.Статьи на русском языке
3.2.Англоязычные статьи
4.ФОРУМ SQL.RU
4.1.Самые популярные темы недели
4.2.Вопросы остались без ответа
 

 

СЕМИНАР

Microsoft Business Intelligence в деталях для ИТ - профессионалов

Узнайте в подробностях как работают решения Microsoft Business Intelligence

Microsoft Business Intelligence - это единый набор интегрированных продуктов, обеспечивающих надёжный доступ к широко распространённым приложениям и отчётам. Основанный на масштабируемой платформе SQL Server, Microsoft Business Intelligence предоставляет информацию пользователям с помощью знакомых инструментов Microsoft Office Suite с использованием возможностей широкого распространения и совместной работы через прозрачную интеграцию с Microsoft Office Sharepoint Server.
Не пропустите возможность узнать о внутреннем устройстве этих продуктов, освоить приёмы работы и получить из первых рук рекомендации экспертов продуктовых команд Microsoft - людей, знающих эти продукты лучше всех. Также обратите внимание на новые решения для управления эффективностью бизнеса, предоставляемые Office Business Applications, и как они могут интегрироваться с существующей ИТ-инфраструктурой вашей компании.

Продукты: SQL Server, SharePoint Portal Server.

Рекомендуемая аудитория: ИТ Директор, Специалист по ИТ

Дата проведения: 02 марта 2007г. 10:00 — 02 марта 2007г. 18:00 (GMT+03:00) Москва.

Время начала регистрации: 09:00

Адрес: Центральный Дом Литератора, ул. Большая Никитская, д. 53, Москва, Россия

Код мероприятия: 1032329368

Докладчики:

  • Marin Bezic - Microsoft EMEA, программ - менеджер SQL Business Intelligence.

  • Kristina Kerr - Microsoft, программ - менеджер Office Business Applications group.

Программа семинара

09:00 ¬ 10:00 Регистрация
10:00 - 11.15 Обзор технологии Microsoft Business Intelligence SQL Server Business Intelligence и Office Business Intelligence Обзор всех компонентов Microsoft Business Intelligence и архитектуры решения. Демонстрация Microsoft BI в действии.
11:30 - 12.45 Создание высокопроизводительных хранилищ данных с использованием Integration Services и Analysis Services Integration Services, Analysis Services Как применить Integration Services для решения масштабных задач по интеграции и трансформации данных. Практические рекомендации по переходу на SQL Server 2005 Analysis Services, новые возможности управления в текущей версии и другое.
12.45 - 13.30 Обед
13.30 - 14.45 Создание системы корпоративной отчетности с помощью SQL Server Reporting Services Reporting Services Практические рекомендации по переходу на новую версию. Как создать сервер отчётов с высокой надёжностью и производительностью. Мониторинг системы корпоративной отчетности в условиях промышленной эксплуатации. Интересные особенности и приёмы работы при использовании системы отчетностив в корпоративной среде.
15.00 - 16.15 Возможности Business Intelligence в Office 2007 Excel, Sharepoint/Excel Services Обзор новых возможностей Business Intelligence и архитектуры решений Office Business Intelligence. Как Sharepoint, Excel Services и Excel взаимодействуют с Analysis Services и Reporting Services.
16.30 - 17.45 Проектирование и развертывание приложений для управления эффективностью бизнеса с помощью Office OBA BSM, Proclarity, PerformancePoint Server Построение и управление системой показателей с помощью Business Scorecard Manager, использование расширенных возможностей для аналитики в Proclarity Analytics Server. Полезные советы по управлению и развёртыванию приложений для управления эффективностью бизнеса с применением PerformancePoint Server (выпуск запланирован на лето 2007 года).
17.45 - 18.00 Сессия вопросов и ответов.

Вопросы относительно участия в мероприятии можно адресовать в Информационный центр Microsoft Россия:
8-800-200-8001 (бесплатные звонки из городов России)
7 (495) 916-71-10, 916-71-71 (для звонков из Москвы)
Факс: 7 (495) 916-71-12
Электронная почта: russia@microsoft.com

СТАТЬИ

Основы ввода-вывода Microsoft SQL Server. Часть 2 (продолжение)

По материалам статьи Bob Dorr: Microsoft SQL Server I/O Basics. Chapter 2

Восстановление на уровне страниц

В SQL Server 2005 Enterprise Edition добавлена возможность восстановления на уровне отдельных страницы, что позволяет осуществлять восстановление поврежденных страниц из резервной копии. Для базы данных можно восстановить из резервной копии одну страницу, не выполняя при этом полного восстановления базы данных, файловой группы, или файла данных. Для получения дополнительной информации, см. SQL Server 2005 Books Online.

Доступность базы данных в фазе отмены (Undo)

В SQL Server 2005 Enterprise Edition доступу к базе данных возможен сразу после завершения фазы повтора (Redo) стадии регенерации (recovery). В фазе отмены применяются механизмы блокировок, которые призваны защитить откат операций. Для сокращения времени простоя, можно объединить восстановление на уровне страниц с послеаварийной регенерацией, что позволяет предоставить доступ к базе данных уже на фазе отмены.

Защита от оборванных страниц (Torn page)

Защита от оборванных страницы не претерпела значительных изменений со времён SQL Server 7.0 и SQL Server 2000. В этой главе подробно описана работа защиты от оборванных страниц, что должно Вам помочь понять разницу между этой защитой, и защитой с помощью контрольной суммы. Наличие оборванной страницы обычно указывает на то, что один или несколько секторов были повреждены.

Распространённые причины

Здесь мы рассмотрим несколько наиболее распространённых проблем, приводящих к появлению ошибки обнаружения оборванной страницы, с которыми сталкивалась служба поддержки Microsoft SQL Server:

  • Подсистемы или аппаратные средства неправильно обрабатывают данные, возвращая перемешанные сектора. Такое наблюдалось у ряда контроллеров или их встроенного программного обеспечения из-за включённого для аппаратного кэша упреждающего чтения.

  • Отключение питающего напряжения.

  • Отражение бит (bit flips) или иные повреждения заголовка страницы. В таких случаях статус страницы указывает на то, что она была оборвана, хотя на самом деле этого не происходило.

Реализация

Защита от оборванных страниц основана на переключениях состояний двух бит из 01 в 10 и обратно при каждой операции записи страницы на диск. Если операция записи А установит биты защиты в 01, значит операция записи B должна установить 10. После этого, операция C должна устанавливать 01 и так далее. Последние два бита каждого 512Б сектора сохраняются в заголовке страницы и заменяются битами оборванных страниц, т.е. значениями 01 или 10.
Соответствующие участники в заголовке страниц данных SQL Server представлены в следующем списке вместе с диаграммой обслуживания бит оборванных страниц.

Участник

Описание

m_flagBits

Битовое поле, которое индицирует состояния: TORN, CHECKSUM или NONE

m_tornBits

Содержит данные для проверки состояний: TORN или CHECKSUM.


Рисунок 1

Биты состояния оборванных страниц устанавливаются в значения 01 или 10, и эти значения хранит участника m_tornBits. Для обслуживания пятнадцати 512Б секторов, каждая пара последних бит этих секторов располагаются с увеличением позиции разрядов, и всё это хранится участником m_tornBits.
Ниже представлена расшифровка шагов, показанных на предыдущем рисунке:

    Шаг 1: В m_tornBits сохраняются значения бит секторов от младших до старших разрядов (подобно небольшому массиву), с увеличением позиций сохраняемых разрядов в соответствии с размещением секторов на странице.
    Шаг 2: В двух последних битах сектора сохраняется текущее значение битов оборванных страниц, которые заменяют имеющиеся там значения.

Биты оборванных страниц ревизуются в момент чтения страницы с диска, если для базы данных включена защита PAGE_VERIFY.

    Шаг 1: Последние биты массива m_tornBits сверяются с образцами (10 или 01), что позволяет убедиться в том, что заголовок не поврежден.
    Шаг 2: Последние пары бит каждого сектора сверяются с соответствующими битами оборванных страниц массива m_tornBits.
    Если какая либо из этих проверок выявит расхождение, страница считается оборванной. В SQL Server 2000 возвратит ошибку 823, а SQL Server 2005 выдаст ошибку 824.
    Шаг 3:

      SQL Server 2000: Первоначальные значения заменяются во время проверки каждого сектора, даже если было обнаружено расхождение. Это не позволяет определить, какой именно сектор был оборван.
      SQL Server 2005: Обнаружение источника проблем усовершенствовано тем, что после обнаружения ошибки биты остаются неизменными. Можно изучить данные на странице, чтобы точнее определить состояние оборванной страницы.

ПРОДОЛЖЕНИЕ СЛЕДУЕТ

[В начало]

ССЫЛКИ НА СТАТЬИ

Статьи на русском языке

Перемещение баз данных сервера SQL Server в новое местоположение с помощью операций Detach и Attach
Microsoft
В статье описывается, как изменить местоположение файлов журналов и баз данных серверов SQL Server 7.0, SQL Server 2000 и SQL Server 2005.

Сервисы офф-лайн синхронизации с SQL Server. Хотите потестировать и пообщаться с разработчиками?
Алексей Халяко
Вышла тестовая версия Microsoft Synchronization Services for ADO.NET http://www.microsoft.com/downloads/details.aspx?FamilyId=75FEF59F-1B5E-49BC-A21A-9EF4F34DE6FC&displaylang=en

Архитектурный манифест: Переход на сервис-ориентированную архитектуру
Микко Контио
В прошлом месяце в этой колонке я говорил о том, что переход к сервис-ориентированной архитектуре - это прекрасный способ работать с традиционным кодом, от которого зависит ваша система, хотя он в значительной степени устарел. По существу, независимо от того, имеете ли вы дело с программой-"бомбой" или просто со сложными требованиями неоднородной системы, переход на СОА чаще всего окажется правильным ходом - а то и единственно возможным - с точки зрения архитектуры. В этом месяце я раскрою преимущества перехода на сервис-ориентированную архитектуру и расскажу, как понять, отвечают ли эти преимущества потребностям вашего бизнеса. Затем я на примере продемонстрирую вам пошаговый план миграции.

По следам Платформы 2007: Visual Studio 2005 Tips & Tricks
Гайдар Магдануров
В этом сообщении я собрал ответы на вопросы, так или иначе связанные с IDE Visual Studio 2005. Советов получилось 10, число ровное, не, не поверите, ничего не подгонял. Все как спрашивали.

Трюки в редакторе Visual Studio
Гайдар Магдануров
После публикации предыдущего поста Visual Studio 2005 Tips & Tricks я подумал, что имеет смысл рассказать о некоторых полезных функциях редактора Visual Studio, о которых порой не знают даже опытные программисты. Привожу несколько наиболее полезных советов.

Регулярные выражения облегчают сопоставление шаблонов и извлечение данных
Дэвид Банистер
Несмотря на то что T-SQL является исключительно мощным языком обработки данных, он плохо приспособлен для анализа текста и манипуляций с ним. Попытки проведения анализа сколько-нибудь сложного текста с помощью встроенных строковых функций приводят к существенному утяжелению этих функций и сохраненных процедур, вследствие чего их становится трудно отлаживать и поддерживать. Но, может быть, существует более подходящий путь?

[В начало]

Англоязычные статьи

Distinct values in SSIS
Alberto Ferrari
Distinct is a partially blocking component that remove duplicates from one flow. Its main advantages against the sort component provided in SSIS are:

SQL Server Books Online
Buck Woody
Most of us have stacks of books on our shelves about SQL Server. But you also have an amazing resource that you get for free with SQL Server, or even on the web: Books Online. Some of you might not know how SQL Server Books Online is put together and what goes into it, so I thought I'd explain what's going on in that part of Microsoft. I'm only talking generally about Books Online here - there are dozens of other teams dedicated to producing and releasing all kinds of documentation and information here at Microsoft, including marketing, legal and many technical teams. Although I'll mention some of the core teams invloved in Books Online, there are many others that are involved in the process. And this kind of work goes on for the server and application teams as well as games and hardware that Microsoft produces. Keep in mind that this is information I've created from my vantage point, which certainly isn't all-inclusive!

SYSK 287: Custom C#/VB (CLR) Types Used in SQL Server
Irena Kennedy
Did you know that you can use custom CLR types in Microsoft SQL Server by adding [Serializable, Microsoft.SqlServer.Server.SqlUserDefinedType] attribute? And, yes, you can implement and use methods - see Microsoft.SqlServer.Server.SqlMethod attribute…

MSDTC Considered Harmful
Matt
MSDTC creates transactions between two database servers, maybe running on different machines, maybe running on different database engines. That way when you mark down Al's account in one server for $10 and mark up Bill's account in the other server for an increase of $10, we don't end up in weird situations where Bill has $10 and Al still has $10, or Bill has $0 and Al has 0$.

SQL Server 2005 Compact Edition Downloads - Honest, it is for real this time (3rd times a charm)
Jim Wilson
Ok… It looks like it's finally here. The false-start in December where the SQL Server 2005 Compact Edition folks pulled the downloads has past. Also, me jumping the gun last week and posting the download URLs before the "official" release is over.

Converting SSIS Data Types to .Net Data Types
Julian Kuiters
If you use Scripting tasks, the CLR or build components for SSIS, you will need to convert values from their SSIS types into .Net types. Here's a quick list of the SSIS data types and their .Net companion.

Shrinking Databases bit-by-bit to avoid long delays
Julian-Kuiters
Michael Jones has a neat script for shrinking databases in small chunks to avoid having the server locked up for a long period.

SQL Server 2005 XML Methods, Part Three, The XPath Expression
Matija Lah
We've discussed the basics of XML in SQL, we've seen a few examples of XQueries and their performance; I guess we're ready to go a bit deeper. In this article we'll look more closely at XPath expressions and see how they behave in SQL Server 2005 XML Queries.

Return related values as array
Matija Lah
Silly it may be but sometimes de-normalized relationships make more sense to the end-user. The following function presents a way to list related records (in the Northwind database) in a flat table, where in this case the names of products in an order can be listed in a single column as comma-delimited values. Also known as: aggregate concatenation.

SQL Server 2005 XML Methods, Part One, Basic Retrieval
Matija Lah
As mentioned in one of the previous posts XML became a native data type with SQL Server 2005. Opinions on its usefulness in a relational database are divided - "fragmented", even. ;) The purpose of this series of posts is to shed some light on how the XML data type could be utilized efficiently, and what are the caveats that should be avoided. In this first part specifically, we'll take a look at the built-in data retrieval methods concerning the XML data type.

SQL 2005 and tempdb
Matija Lah
With the expansion of the row versioning framework in SQL Server 2005 tempdb becomes an even more important part of the platform than it used to be. Snapshot transaction isolation as well as database snapshots make big use of this "temporary" storage facility and keeping track of its usage is the key to the continuous efficiency of data management solutions based on the SQL Server platform.

Tame Those Strings - Finding Carriage Returns
Steve Jones
Expanding on his series of string manipulation in T-SQL, Steve Jones takes a look at how you go about removing those unseen characters from your strings.

Why RAID is (usually) a Terrible Idea
Jon Bach
As president of Puget Custom Computers, I get a unique perspective on computer products and technology. Our company specializes in selling high performance custom computers, and that naturally brings up the question of RAID often. There is an overwhelming opinion out there that if you have the money and want a blazing fast and stable computer, that you should put your hard drives in RAID. We have known for years that this perception is just flat out wrong, but the problem is that the idea is so widely accepted that it is nearly impossible to convince our customers otherwise. In fact, if we try too hard to talk them out of it, we end up losing the sale! So, should we be selling configurations that we know are flawed, for the sake of making the sale? To be honest, if it comes down to it we will, but not without a fight! This article is just the latest effort in educating the public about RAID.

Data Driven Subscriptions for Reporting Services (2000 and 2005)
Jason Selburg
Reporting Services has proven to be one of the more useful subsystems of SQL Server. However the standard editions of both SQL Server 2000 and SQL Server 2005 have limitations with regard to the subscription features. Jason Selberg brings us an updated version of his code for extending the subscription feature.

How to setup Database Mirroring in SQL Server 2005 using T-SQL
Tara Kizer
Our disaster recovery and database availability solution involves clustering and Database Mirroring in SQL Server 2005. At our primary site and our disaster recovery site, we have a 4 node cluster with several SQL Server 2005 instances on it. So that's 8 nodes and 2 clusters at both sites. The cluster is Active/Active/Active/Active at the primary site.

See performance gains by using indexed views in SQL Server
Tim Chapman
When you use index views in the right situations, they can dramatically improve the performance of SQL Server queries. Tim Chapman explains why performance testing is the best way to tell if indexed views will be beneficial to your database.

Entity Client
Zlatko Michailov
With the next release of the .NET Framework, ADO.NET will introduce a new member of the family of data providers - EntityClient. The mission of EntityClient is to provide a gateway for entity-level queries. Through EntityClient one queries against a conceptual model, not against a specific store implementation of that model. EntityClient does not directly communicate with the data store but it requires a separate, store-specific, provider. EntityClient can work in conjunction with SqlClient that come in the ADO.NET bundle. Additionally, we have launched a partner program to help third-parties produce compatible store providers.

Counting number of records for all the tables - SQL Server
Decipherinfosys
Here is a quick and dirty way of getting the record counts from a database schema in SQL Server (works on both SQL Server 2000 and SQL Server 2005):

Collecting nightly statistics in SQL Server
Decipherinfosys
As promised in one of the earlier blog posts, here is a stored procedure that you can use to collect stats for the tables and their indexes in SQL Server. You can pass in a number as the percentage for the percent of the data that should be sampled for collection of the stats. Ideally, if you do have a window of opportunity, you should collect full 100% stats. Though, in large systems, this will not be possible because of the amount of data that resides in those databases. In addition, you should have the "auto create statistics" and "auto update statistics" option on for the database. If you do have a system which is a true OLTP system and has a lot of transacational load, you might consider turning "auto update statistics" to be off since you do not want the optimizer to kick in (between the transactional operations ) to start collecting the stats. That can also cause query recompilations. Optimizer does that based on a formula that is detailed in one of the Microsoft KB articles and is loosely based on 700 or so DML operations against the table/index.

ANSI_PADDING and trailing spaces in SQL Server
Decipherinfosys
In SQL Server, ANSI_PADDING setting controls the "storage" part of the trailing space and that is done at the time of the creation of the tables. SQL Server follows the ANSI standard on how to compare strings with spaces. The ANSI standard requires padding for the character strings used in comparisons so that their lengths match before comparing them. The padding directly affects the semantics of WHERE and HAVING clause predicates and other Transact-SQL string comparisons. For example, Transact-SQL considers the strings 'xyz' and 'xyz ' to be equivalent for most comparison operations.

Querying the total number of records, but only returning a page of them
Rob Farley
A lot of people seem to have been asking about this recently. They have a T-SQL query which uses a call to row_number() within a CTE or derived table, and uses this row_number() field to filter the results to a single page - but they also want to know how many rows were in the whole set (rather than just in the page). @@rowcount doesn't do it for them, because this will return the number of rows in that page.

Managing SQL Server using PowerSMO
Dan Sullivan
PowerSMO combines two new Microsoft technologies:

Pop Rivett and the Uncontrolled Release
Pop Rivett
Pop Rivett gazed reproachfully at Clarke Minor. 'I ended up teaching Needlework to the infants today dy' know, Clarke Minor. Dashed embarrassing for a Technology teacher, what, given that it is not really my subject'

How to Downgrade a Database from SQL Server 2005 to SQL Server 2000
Panos Tzirakis & George Dounavis
As you may all know, SQL Server 2005 request a minimum of 8GB RAM to work… let say satisfactorily. I first didn't knew that and after a while from the upgrade I did from SQL Server 2000 to 2005 my SQL Services were starting to crash three or four times per DAY!!!

Is NULL So Bad?
Mystery Blogger
I ran across an article online by Alf Pedersen (http://www.databasedesign-resource.com/null-values-in-a-database.html) entitled "NULL values in a database: Should it be allowed?" and upon reading it, made me realize why NULL is such a "hot topic" Let me explain…

Solution to February Reader Challenge: Check for an Empty Table
Umachandar Jayachandran
There were no winners for the February Reader Challenge, "Check for an Empty Table." Here's a recap of the problem and the solution to the February Reader Challenge.

Gartner's BI Report is Magically Interesting
Brian Moran
At the end of January, the Gartner Group released its "Magic Quadrant for Business Intelligence Platforms, 1Q07" report. Last September, in "The Magic of BI" ( http://www.sqlmag.com/Article/ArticleID/93683/sql_server_93683.html ), I wrote about Gartner's "Magic Quadrant for Data Warehouse Database Management Systems 2006" report, which was similar to the current report and overlaps product coverage in some ways. But as their names suggest, the two reports are different. The data warehouse report released last fall focuses more on BI from a data warehouse perspective, whereas the report Gartner issued a few weeks ago focused more on "pure" BI. I think it was Shakespeare who said "BI by any other name is still BI" (they had dimensional modeling in the time of the bard, right?). I could devote multiple articles to simply parsing the differences between the terms BI and data warehousing, but in the interest of time and space, let's just say that you should read both of the reports if BI or data warehousing is a topic that's in your future. Are you a database professional? Then both topics are in your future, whether you realize it or not.

Does SQL Server 2005 Clustering Really Make Sense for My Organization's Needs?
Brad M. McGehee
As the forum administrator for SQL-Server-Performance.Com, I see hundreds of questions from DBAs on SQL Server clustering. What always surprises me is how misunderstood clustering is. So exactly what do I, and Microsoft, mean when we refer to SQL Server 2005 clustering? Here's a good working definition:

The easy fix to mysterious SQL collation errors
Jonathan Rabson
I'm doing this query that's the same SELECT statement for two tables with a UNION ALL connecting them. While each query works great on its own, when I try to run them together (with the UNION ALL) I get this error message:

Practical Business Intelligence with SQL Server 2005: Data Mining
John C. Hancock, Roger Toren
This sample chapter looks at how to use some of the data mining features in SQL Server Analysis Services 2005 to perform tasks such as customer segmentation and market basket analysis. The data mining results are presented in the form of new dimensions in cubes and are used in Web applications.

Predicting based on rules alone and getting everything you always wanted
Jamie MacLennan
In this post I explained how to couch an associative prediction that omitted predictions that were based solely on marginal statistics. The one catch was that if you asked for 5 items, you might not get 5, you could get 4 or 3 or even down to 0, since all predictions without a corresponding prediction node id were filtered out.

Predicting the non-majority state
Jamie MacLennan
The Predict() function in DMX always returns the highest probability non-missing (by default) state for the requested attribute. That is, if there are two states, the state with a probability higher than 50% will be returned. However in many cases you want a response to be a target state even though it may not have the highest probability. For instance, say you implemented the AdventureWorks Bike Buyer model, and your profit chart indicates you will recieve the highest profit by selecting all the cases where the probability of "Buy" is greater than 25%. Given the processes in place, you don't want to have a filtered result, rather you want to label each row in a table as a target or not.

To Synchronize or To Refresh?
Gert Drapers
Are you confused too? Are you wondering what the difference is between the "Refresh" button in "Solution Explorer" and the "Synchronize" button in "Schema View"? Both buttons share the same icon; but they are named differently; are both option doing the same or different things?

Zero to high availability in 60 seconds...
Matt Hollingsworth
I am a car fanatic. My first car was a Triumph GT6+ which I noticed under a tarp in my neighbors garage. It had been sitting idle for ten years. Six years ago, I modified a Mustang extensively, adding Brembo brakes and a supercharger, exhaust headers, a modified suspension, etc and it will accelerate from zero to sixty miles per hour in just over five seconds. When new car magazines come out, I pore over new car statistics, constantly in search of exciting cars across the performance, style, price and other measures. Sometimes friends will ask me which car I like the best. I can never answer this question because each car has a different personality based on the trade off decisions made during the design of the car. The recent ice storm in Seattle makes the point well - the Mustang was completely immobile on the ice.

Database virtualization is virtually certain
Matt Hollingsworth
You may not realize it, but operating system virtualization is at least 40 years old. That's eons in the technology world. The first virtualization offering that I remember was IBM VM, however there were others before that point. IBM had an internal project around 1966 that eventually surfaced as VM, but not before IBM released MVS (Multiple Virtual System), which was a partially virtualized operating system for the mainframe. MVS/ESA was the first operating environment I used as a programmer upon graduation. Why didn't they release the original VM version of the OS before MVS? The rumor was that there were concerns about how to license the technology in a way that allowed them to recapture the development costs of the operating environment. For more information, check out http://en.wikipedia.org/wiki/VM/CMS

When inserts don't scale, it's not time to bail
Matt Hollingsworth
I often encounter questions on why an insert heavy workload doesn't scale well when moving a workload from a small server in a test environment to a much more powerful production server. Usually customers resolve the issue by adding multiple parallel insert processes or enlarging the size of the transactions. The reason that this is true is tied to how the database writes to the log. To maintain the reliability of data, the database must ensure that the log is written to disk before the database returns a success from the commit - this covers the case of a power failure. When the application issues the commit, the modified data buffers will be written out to disk sometime in the future, but the number one priority is to flush the log buffers from the transaction to disk. Once the data is confirmed by the OS as having been written to disk, the application will receive the successful message and continue. If a power outage occurs between the commit and the response, all is well because the application never received a successful response. The database ensures the data is either committed or not.

How safe is your database?
Matt Hollingsworth
Ensuring your database is protected from loss is about as enjoyable as a dentist visit. We all know that we should have backups, and maybe even have thought about where we store the backups. I want to mention two items to think about today to save you pain later. The first is obvious when you think about it, but you may not have time to put it in perspective during your normal day.

Got database scale?
Matt Hollingsworth
As you might expect given the size of Microsoft, there are many websites which must be able to scale to very high levels. Have you ever wondered how this is acheived? It would be impossible to use a single server to support the databases on the highest volume sites. Solutions for scaling across multiple servers typically revolve around two approaches which are sometimes combined:

When you need to fix copies of data that get out of sync with each other
Matt Hollingsworth
I can't tell you how many times I've been in the situation of trying to figure out what has changed between two tables. Often this happens when you create copies of schemas. Maybe you have a dev, test, QA and production environment and they all contain a reference table (for example sales tax by state) but the table has been updated over time. How do you tell that the data between the environments is different? How do you tell which data has diverged? How do you bring it back in sync?

Using SAN Volumes in Project REAL
Performance Team
As we started Project REAL, we were fortunate to be able to use multiple machines and SAN devices to test different configurations. At the same time, this presented a challenge in that we needed to keep track of what information was on what SAN volumes even as those volumes could be switched from one machine to another. This note describes some techniques we learned for managing the various SAN volumes.

Storage Modes in Analysis Services
Len Wyatt
Why did we only use the MOLAP storage mode in Project REAL? There is also ROLAP and HOLAP storage, but they weren't used at all. Also, why doesn't Project REAL illustrate pro-active caching?

High Volume Update Performance
Robin
SQL Server does a great job of optimizing processing across a wide range of runtime scenarios, while also providing a flexible configuration and execution environment. One of the tradeoffs associated with this level of flexibility is that different configurations and implementations, while logically equivalent in terms of results, can yield very different performance.

Tools of the Trade: Part I - Universal Data Link Files
Justin Trobec
The other day I was working on an issue with one of our customers.

PART II - Data security enhancements in SQL Server 2005
Sergey Pustovit
In the previous edition (Data security enhancements in SQL Server 2005) , we talked about data encryption/decryption in general.

Passing lists to SQL Server 2005 with XML Parameters
Jon Galloway
SQL Server 2005's XML capabilities make it a easier to pass lists to SQL Server procedures.

Synchronization Services for ADO.NET Vs SQL Merge Replication
Rafik Robeal
One of the common questions I get from developers once they get their hands on the new sync framework for ADO.NET is: How is it different from SQL Merge Replication? When should I use merge as oppose to sync services? Can sync services do filtering, ddl replication etc …? These are all interesting questions, so let's talk about the scope of each product:

Support for RS 2000 RDL in RS 2005
Lukasz Pawlowski
I've started getting questions like the following:

Windows SharePoint Services and RS 2005
Lukasz Pawlowski
In RS 2000, users had a lot of grief to get RS to work on the same computer as Windows SharePoint Services (WSS). The reason for this is that WSS version 2.0 effectively commandeers the computer. We provide a trouble shooting document for these here: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/RSinstall/htm/gs_installingrs_v1_9fdy.asp

Monitoring Subscription Status (also calling RS SOAP methods from inside reports)
Lukasz Pawlowski
While posting on the Reporting Services forum, I came across a post by some folks looking to know how to automatically find out that a subscription had failed to run.

Populating the Cache before your users pound your report server
Lukasz Pawlowski
Often people take a look at their report execution times and wonder how to optimize it. They readily see that it is possible to cache a report the first time a user runs it so that subsequent users don't have to incur the initial execution time.

Upgrading Reporting Services - What about my Report Definitions (RDLs)?
Lukasz Pawlowski
Some folks have asked me what happens when I upgrade my RS 2000 to an RS 2005 deployment with respect to Report Definitions (RDLs). This is an interesting lifecycle problem that we've approached using the following idea:

SQL Server 2005 SP1 Reporting Services database expected version is 'C.0.8.43' problem
Lukasz Pawlowski
(Originally I thought this post should be titled: SQL 2005 SP1 Tattered my Report Server Database - What's up with that?)

Per user snapshots? Or multiple snapshots for a report each with different parameters at the same time
Lukasz Pawlowski
I've received questions in the past about whether the report server supports per user snapshost - that is report snapshots customized with different parameter values for a given user. The answer I have always given is no - you get one snapshot per report and that snapshot always uses the default parameter values.

How to diagnose issues when running reports in the report server?
Lukasz Pawlowski
A customer came to the SQL Server Customer lab to investigate a problem with one of the feature I'm responsible for - data-driven subscriptions. In looking at their solution, we discussed any number of problems they had encountered. I realized that it can be difficult to find out where to start looking when a problem occurs. This post will hopefully provide you a starting point as you endeavor to fix the issues you run into.

Queries On Steroids ... Part IV
Danny Lesandrini
Query analysis tools for the stout of heart. Let me begin by confessing that I am not qualified to write this article. Everything you read below comes from Internet research, primarily from two sources:

Recovery Planning For SQL Reporting Services
Mike Pearson
SQL Server 2000 Reporting Services is becoming a more and more popular reporting option everyday. However, the disaster recovery plan for this add on is not a simple backup and restore since there are multiple pieces and servers usually involved. However the DBA may be responsible for the entire system. Mike Pearson brings us a look at some of the scenarios that you need to consider and what you might need to prepare for disaster recovery of SQL Server 2000 Reporting Services.

NUMA and Sqlserver
Saurabh Singhal
Recently there was an issue in our labs related to NUMA and what is treated as NUMA by SqlServer2005. Here are some findings:

SQL Server on VM
Rick Heiges
Recently, someone asked me a question about consolidating SQL Server to VMWare. Here is my response.

xsi:type magic
Denis Ruckebusch
As I said in my previous post, the past few weeks have been demanding and I had little time to spend on blog post. However I recently rediscovered an interesting behavior in typed XML DML and I felt I should make some time for a quick post.

Complementing XSD with CHECK constraints
Denis Ruckebusch
I was recently asked to look at a customer's question. This person wanted to write a schema that would validate instances like the following one.

xsi:nil magic (part 1/2)
Denis Ruckebusch
A few months ago I wrote a post about the interesting behavior of the xsi:type attribute. Today we're going to look at his no so distant relative xsi:nil.

xsi:nil magic (part 2/2)
Denis Ruckebusch
Last time we looked at what happens to the xsi:nil attributes when replacing the value of a simply typed element. In this post, we're going to look at complex types.

Mixed content (part 1 of many)
Denis Ruckebusch
I started working on this two days ago. I quickly realized that when I'm done I'll most likely have material for multiple posts. My original plan was to finish the article in the next few days, and then publish it as several posts over time. However I eventually decided to start posting now, while the work is still in progress, and let any feedback or questions you might have guide me.

Mixed content (part 2)
Denis Ruckebusch
Following last week's introduction to the concept of mixed content, let's look in detail at a validated XML instance with mixed content.

High Volume Update Performance Benchmarks
Glenn Berry
Microsoft's SQL Server Performance Blog has a very interesting post today about different methods for updating several VARCHAR columns and one DATETIME column on a 500 million row table, using a very beefy 16 CPU server with 128GB of RAM. It turns out that doing an INSERT into a temporary working table on another filegroup (not a TEMP table), helps relieve the I/O contention, and reduces the time for the operation by 87%.

An Introduction to SQL Server 2005 Clustering Basics
Brad M. McGehee
Clustering is a complex technology with lots of messy details. To make it easier to understand, let's take a look at the big picture of how clustering works. In this article we take a look at:

[В начало]

ФОРУМ SQL.RU

Самые популярные темы недели

Вышел русский SQL Server 2005!!!
Суррогатные или естественные
Секционировать или нет?
Размещение базы данных на физически разных серверах
Как быть с текущей датой???
Перебор и обработка записи в хп, триггере...
MD5,SHA1 на T-SQL
Импорт/экспорт таблиц. Перенос данных таблиц с одной БД в другую
Знак вопроса в имени процедуры...
Разговор на кухне / Зависания MS SQL
Пересечение временных промежутков
Табличная переменная или n-запросов ?
от новичка: КАК СДЕЛАТЬ чобы хранимка ВОЗВРАШАЛА КОЛ-ВО вставленных ЗАПИСЕЙ?
Помогите с запросом.
Ошибка SQL. 08S01. Общая ошибка сети.
багофича FMTONLY
Как заставить SQL сервер не отдавать память системе?
Проблема с несколькими ADO клиентами
"Провалы" производительности на BULK INSERT
Жуткие Тормоза в 1С на SQL Server

[В начало]

Вопросы остались без ответа

Обмен данными VFP-SQL
Опции setup для 2005 EXPRESS
Проблемы с Report Services
MSDE SP4 окромя MS
Не создается maintenance plan
В прекрасный момент пропали файлы с расширением mdf и lck
проблема с перекачкой данных из *сsv в Бд
Как изменить Server collation
Как подключиться к SQL Server 2005
Права, права...
Помогите с запросом.
distinct и паралелизм

[В начало]

#346<<  #347  >>#348
Rambler's Top100 Рейтинг@Mail.ru  Administrator: Обратная связь 
Copyright: SQL.Ru 2000-2013