Press "Enter" to skip to content

Author: Kevin Feasel

Tips for Using Azure Table Storage

Published 2020-10-22 by Kevin Feasel

Adrian Hills takes us through using Azure Table Storage:

Azure Table Storage is a NoSQL key-value PaaS data store that can be a great option for highly scalable, highly available systems. It supports storing petabytes of data and a flexible data schema, meaning different entities in the same table can have different schemas. References to NoSQL databases having “flexible schema” or being “schema-less” can give the impression that database schema design is a thing of the past and that you can bypass it and focus more on the application code. The reality is, even in this NoSQL world, schema design is very important and if you don’t give it due care and attention, then it can come back to bite you.

If you have a RDBMS background and are new to Azure Table Storage, it’s common to find yourself “thinking in SQL” and trying to solve database modeling requirements with a SQL approach before then trying to translate that to a key-value mindset. In this blog post, I’ll cover some of the fundamentals of Azure Table Storage and dive into some common questions you might find yourself asking about Azure Table Storage. Where code samples or references are applicable in this blog post, we’ll be focusing on .NET and using the Azure SDK (specifically relating to the Microsoft.Azure.Cosmos.Table nuget package).

Read on for the full story.

Comments closed

Importing Perfmon Data Into SQL Server

Published 2020-10-22 by Kevin Feasel

David Klee continues a video series:

If you followed the last video, I showed you how to set up windows perfmon for ongoing performance metric collection on all of your critical SQL Server machines. However, having this raw data in a portable format doesn’t mean you have an easy means to access that data. In this training video, I show you a PowerShell script that we released that helps you export this data into a SQL Server database table so that you have access to the raw data to perform ongoing performance analysis with whatever favorite tool you prefer.

David walks through the process in a video, so check it out.

Comments closed

Self-Service with Azure Synapse Analytics

Published 2020-10-21 by Kevin Feasel

Paul Andrew lays out an interesting idea:

I’ve been playing around with Azure Synapse Analytics for a while now exploring the preview features and trying to find a meaningful use case for the ‘single pane of glass’ capabilities. In this post I’m exploring one possible option/idea for creating a very simple self service approach to dataset ingestion and consumption. Full disclosure, the below is far from technical perfection for lots of reasons, I mainly wanted to put something out there as an idea and use it to maybe start a conversation.

Click through to see Paul’s take on the matter.

Comments closed

Choosing the Right Index and Partition in Dedicated SQL Pools

Published 2020-10-21 by Kevin Feasel

Tsuyoshi Matsuzaki gives us some advice on indexing and partitioning data in Azure Synapse Analytics dedicated SQL pools:

Designing index for a table is so primitive and important for better performance.
There’s no “one answer for any case”. You should choose right index for a table depending on the size, usage, query patterns, and cardinality.

In order to help you understand pros/cons in each indexes, I’ll show you each pictures illustrating intuitive structures of indexes available in Synapse Analytics.

Because dedicated SQL pools aren’t the same as the SQL Server box product, it’s important to go in with the understanding that indexing won’t be exactly the same as on-premises or in Azure SQL Database.

Comments closed

The Raw Facts on Azure SQL DB Serverless

Published 2020-10-21 by Kevin Feasel

Taiob Ali gives us a briefing summary on Azure SQL Database Serverless:

Occasionally, load balancing automatically occurs if the machine cannot satisfy resource demand within a few minutes. For example, if the resource demand is 4 vCores, but only 2 vCores are available, it may take up to a few minutes to load balance before 4 vCores are provided. The database remains online during load balancing except for a brief period at the end of the operation when connections are dropped.

Click through for more points along these lines.

Comments closed

Dynamic Format Strings when using Calculation Groups

Published 2020-10-21 by Kevin Feasel

Alberto Ferrari shows off how you can dynamically generate format strings when using calculation groups in Power BI:

Each product in Contoso weighs a certain weight. The weight is stored in two columns: the unit of measure and the actual weight, expressed in that unit of measure. Specifically, Contoso uses three units of measure: ounces, pounds, and grams.

Because the units of measure are different, you cannot aggregate the weight over different products. If you author a simple measure that computes the ordered weight of products by using a simple SUMX, the result is wrong:

Click through to see how you can work through this problem.

Comments closed

SQL Server Management Studio 18.7

Published 2020-10-21 by Kevin Feasel

Drew Skwiers-Koballa announces SQL Server Management Studio version 18.7 is now generally available:

Policy-based management is accessed in SQL Server Management Studio under “Management” in the object explorer as “Policy Management”. Getting started with policy-based management can be accelerated by importing the sample policies available for SQL Server. In September, these policies were added to the open source collection of SQL Server samples to facilitate their use and improvement. You can access these sample policies on the GitHub repository and your contributions to these best practices are welcome. For more information on Policy-Based Management, please check out the documentation.

I think Policy-Based Management is one of the biggest missed opportunities in SQL Server. They came out with a good start in 2008 but the product stagnated after that and it remains under-utilized as a result. Perhaps open-sourcing the policies will help, as the key problem with PBM was how limited it was.

Comments closed