sqlsathistory

Nr: 274

Back to Main list

SQLSaturday #274 - Slovenia 2013

Start Time (24h) Speaker Track Title
00:00:00 Boris Hristov DBA Replay your workload as it's your actual one!
00:00:00 Dejan Sarka DEV Optimizing Temporal Queries
00:00:00 Kennie Pontoppidan BI Autogenerating a process data warehouse
00:00:00 Marcin Szeliga DEV Indexing for denormalization
00:00:00 Mark Stacey BI Analysis Services for the DBA
00:00:00 Miloš Radivojević DEV TSQL Performance Tips
00:00:00 Miloš Radivojević DEV SQL Server 2014 - New Cardinality Estimator
00:00:00 Pieter Vanhove DBA Harder, Better, Stronger and more secure
00:00:00 Rasmus Reinholdt BI Optimizing BI with SQL2014 In-Memory tables
00:00:00 Mladen Prajdić DBA From SQL Traces to Extended Events. The next big switch.
00:00:00 Tomaž Kaštrun BI Using Data Quality services on Customer related data
00:00:00 William Durkin DBA SQL Server Consolidation - Resistance is Futile
00:00:00 André Kamman BI Automate your SSIS development with BIML
00:00:00 Marco Parenzan DEV .NET Development for the SQL Server Developer

SessionID: 10903

Replay your workload as it's your actual one!

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DBA

Speaker: Boris Hristov

Title: Replay your workload as it's your actual one!

Abstract:

Have you ever wanted to replay your application workload be that because you upgraded to a new hardware, added new indexes or something else? Now there's a feature for that! It's called Distributed Replay and in this session you will learn how this new feature in SQL Server 2012 can help you resolve that problem!

SessionID: 13515

Optimizing Temporal Queries

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DEV

Speaker: Dejan Sarka

Title: Optimizing Temporal Queries

Abstract:

Having a SQL Server solution for a problem does not mean the job is done. Of course, the next immediate issue is the performance. Temporal queries that involve intervals are typically very IO and CPU intensive. For example, a test for overlapping intervals was solved with inefficient queries for years. However, a handful of solutions with fast queries was developed lately. This high-level technical session introduces five different methods to get efficient queries that search for overlapping intervals. Of course, these solutions can be implemented on other temporal problems as well. Actually, the test for overlapping intervals is one of the most complex temporal problems.

SessionID: 18106

Autogenerating a process data warehouse

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: BI

Speaker: Kennie Pontoppidan

Title: Autogenerating a process data warehouse

Abstract:

In many data warehouses we model fact tables with measures based on attributes which we can count and do arithmetic upon. It is more difficult to handle fact tables with measures based on length of intervals of events happening in the source systems. So questions like "What are the typical waiting times in our order process?" are seldomly modelled in the data warehouse. Especially if the event data comes from different source systems.

In this talk I will show you how Rehfeld has overcome this difficult problem using an autogenerated datawarehouse based on metadata on the objects in the EDW/DM layers. In this approach you can autogenerate the process data warehouse if only you have the timestamps present in the EDW layer and are able to correlate them. The talk will consist of slides and demos.

SessionID: 19450

Indexing for denormalization

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DEV

Speaker: Marcin Szeliga

Title: Indexing for denormalization

Abstract:

What has the biggest impact on SQL Server overall performance? Hardware, SQL Server configuration or maybe query tuning? No, none of those frequently discussed options. They are important, but the single, most important factor is the database design.

A third normal form is great, but sometime a database really should be denormalized to gain a performance boost. This is where indices shine. During this session you will see: how important a proper table design can be, when to use indices on computed columns and why they are way better than triggers, how to create useful indexed views and how to asses their costs and benefits.

SessionID: 20085

Analysis Services for the DBA

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: BI

Speaker: Mark Stacey

Title: Analysis Services for the DBA

Abstract:

Many DBAs have heard the term Analysis Services, but don't really know what it's for or who would use it. In this session, Mark will explain the use cases for Analysis Services, illustrate how it improves performance for large data sets and eases report writing, and will go through a practical example of building a first Analysis Services solution.This session will be practical for SQL 2005 through 2008 R2, but he will also take the time to explain what changes in SQL 2012 and Tabular mode.

SessionID: 20737

TSQL Performance Tips

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DEV

Speaker: Miloš Radivojević

Title: TSQL Performance Tips

Abstract:

SQL Server optimizer doesn't use and index seek for execution of your query although the quer is high selective? What is better, when and why: LIKE vs: SUBSTRING, IN vs. EXISTS, SUBQUERY vs. JOIN. Why you should not use the UPPER or LOWER functions? How to avoid non-SARG-able WHERE clauses? Which query takes about 3 minutes in SQL Server 2008 and only one second in SQL Server 2012?

In this session we will advice how to avoid common developer mistakes and show how bad designed queries reduce the SQL Server optimizer's capabilities for choosing an optimal execution plan.

SessionID: 20738

SQL Server 2014 - New Cardinality Estimator

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DEV

Speaker: Miloš Radivojević

Title: SQL Server 2014 - New Cardinality Estimator

Abstract:

In SQL Server 2014 query processing has been improved. What we can expect from new cardinality estimator? Does it really estimate better? Will our execution plans be better after migration to 2014? All of them? Is the key ascending column statistics problem solved in new version? Do we finally can have more than 200 steps in the statistics histogram per table? What about column correlation and cardinality estimates - are the predicates still completely independent?

This session will answers all these questions and will demonstrate new query processing capabilities and compare them with those from SQL Server2012.

SessionID: 22141

Harder, Better, Stronger and more secure

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DBA

Speaker: Pieter Vanhove

Title: Harder, Better, Stronger and more secure

Abstract:

With natural disasters like earthquakes and fires in the news so often, IT Professionals know that many disasters can shut down a SQL Server.

Microsoft has already implemented a lot of HA/DR solutions in previous editions of SQL Server, like AlwaysOn failover clustering, AlwaysOn Availability Groups, log shipping, ... But,when it comes to high availability, SQL Server 2014 changes the game by enhancing the availability, scalability, and flexibility of AlwaysOn. How about your HA/DR solution? Did your management decide the DR site is still too expensive? Are your backups secure? In this session, you will learn all about the new and enhanced features of AlwaysOn in SQL Server 2014 like managing backups to Azure, encrypting backups, adding

SessionID: 23500

Optimizing BI with SQL2014 In-Memory tables

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: BI

Speaker: Rasmus Reinholdt

Title: Optimizing BI with SQL2014 In-Memory tables

Abstract:

Well - they named it SQL2014 In-Memory OLTP tables - BUT that does not mean we can't optimize our BI solutions with this new feature.

This demo intensiv session is build around a number of hypothesis that are testet and show what we can - and cannot optimize with In-Memeory tables covering both the ETL and (R)OLAP angels of the BI process. The session will use the latest CTP release and is constantly updated to include the latest news on the topic.

SessionID: 24786

From SQL Traces to Extended Events. The next big switch.

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DBA

Speaker: Mladen Prajdić

Title: From SQL Traces to Extended Events. The next big switch.

Abstract:

This session will give you insight into both of the current tracing technologies in SQL Server with emphasis on Extended Events.

The old SQL Trace which has served us well over the past 10 or so years is on its way out because the overhead and details it produces are no longer enough to deal with today's loads. The new Extended Events are a new lightweight tracing mechanism built directly into the SQLOS thus giving us information SQL Trace just couldn't. They were designed and built with performance in mind and it shows. Mastering Extended Events requires learning at least one new skill: XML querying. We'll look at how the Extended Events work with only a brief overview of old SQL Trace architecure.

SessionID: 26999

Using Data Quality services on Customer related data

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: BI

Speaker: Tomaž Kaštrun

Abstract:

Main focus of this session will be usage of data quality services with practical examples on managing quality, integrity and consistency of data for further data usage. Session will show how to build a knowledge base, manage domains and use data cleaning and matching.

SessionID: 28199

SQL Server Consolidation - Resistance is Futile

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DBA

Speaker: William Durkin

Title: SQL Server Consolidation - Resistance is Futile

Abstract:

The IT world is currently in cost-saving mode - cutbacks are all over the place and then Microsoft pushes out SQL Server price hikes.

In this session we will discuss how consolidating SQL Server can reduce overall costs and increase efficiency. We will cover the reasons for consolidation along with the advantages and the disadvantages of consolidating SQL Servers. We will also take a look at some of the tools that can assist with a consolidation project and some pitfalls that I have experienced on previous consolidation projects.

SessionID: 34558

Automate your SSIS development with BIML

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: BI

Speaker: André Kamman

Title: Automate your SSIS development with BIML

Abstract:

Ever tried to import a file with the Import/Export wizard?

Or created a bunch of SSIS packages to process a data warehouse load? Then you know how much work it is to specify the metadata correctly just to create a package that actually works. Wouldn't it be cool if you had a descriptive language which looks at your metadata and just created the packages for you? This is what BIML is all about. In this session I explain what BIML is, how it works and I'll show you how you can generate your packages and quickly respond to changes. You can expect a demo rich session with lots of notes from the field and practical examples. This is not just for BI developers, DBA's or SQL dev's who need to import or export data occasionally will learn some quick and easy tricks as well.

SessionID: 34751

.NET Development for the SQL Server Developer

Back to calendar

Event Date: 21-12-2013 - Session time: 00:00:00 - Track: DEV

Speaker: Marco Parenzan

Title: .NET Development for the SQL Server Developer

Abstract:

Every SQL Server DBA or Developer needs to create some kind of applications: a dashboard for data, a management console, a data-entry tool. Like SQL Server, .NET have evolved over years. Linq and ORM, ASP.NET MVC, Javascript/jQuery, async programming, Azure and the Cloud. I'll present a fast entry to .net and some scenarios and sample applications that a typical SQL Server DBA or Developer can use in his daily work.