We had a great live presentation on November Friday 12th, 2021, at the Data Community Summit 2021. My colleague Gail Zhou, and I, presented a set of feature parities, capabilities and recommendations when migrating a Data Warehouse from SQL Server to Azure Synapse Analytics Provisioned Pool. We also went ahead and provided a hands-on demo on how to a assess the environment, as well as how to migrate objects and data from SQL Server.
In November 2020, for the (until then) paid conference SQL Pass 2020, my colleague Avnish Rastogi and myself, presented the topic of how to migrate hundreds of SQL Server databases to Azure at scale.
We executed as Architects within of the Data Migration group at Microsoft, and worked in several engagements assessing multiple source databases and landing them into Azure. From the use cases, we compiled the steps, methodologies and services required to asses and migrate on-premises SQL Server instances to Azure Managed Instance. Leveraging PowerShell, we were able to automate and scale the migration of multiple databases.
We have been experiencing an issue with a clustered node where instances will run extremely slow and the only process eating CPU resources is “NT Kernel & System”. We have upgraded drivers and even wiped out and reinstalled Windows and SQL Server (both 2008 R2) and the issue reappears.
In the process of troubleshooting with Microsoft, we needed to fail over instances to collect log information. We have done this several times. Because our latest cluster implementations include Active/Active nodes, we need to adjust Max Memory to SQL Server whenever all instances are in one node.
Yesterday something did not go right and a human error caused the instances to failed. Even experienced DBAs will eventually break something, but they will do anything possible to correct. Memory was set to 27 instead of 27000 MB. No connections could be made to the instance.
This blog entry is participating in the T-SQL Tuesday #16 for the month of February, hosted by Jes Schultz Borland (Blog|Twitter).
Back in 2001 I was involved in a project migrating an e-commerce site to a new platform. We were mapping and moving data from an Oracle 8.1.5 to an Oracle 8.1.7 instance. The framework was the beloved Dynamo running on Sun Solaris. The Dynamo engineers decided to use sequencers on each entity (e.g. product, category, etc). I was trying to figure out the best reusable PL/SQL block to move the data creating the correct sequence until I was pointed out that Oracle 8.1.6 supported Analytical Functions. After reading and practicing I discovered the power of those functions.
For the purpose of this blog I will refer the Analytical Functions as Windowed Functions instead. They are very useful as it allows the user to crunch measures of subsets of data maintaining the “raw” detail level data. For example, it allows us to get the average price of a product category while also displaying the unit price of each product. It can also help us rank a product by its price and “window” it by category.
1 2 3 4 5
-- Qty of products and average price SELECT COUNT(productKey)AS[Quantity] ,AVG(StandardCost)AS[AverageCost] FROM[dbo].[DimProduct]
-- Take the Average for each Product Subcategory and attach to the detail level data SELECT AVG(StandardCost)OVER(PARTITIONBY[ProductSubcategoryKey])AS[AverageCostSubPerCat] ,productKey ,[ProductSubcategoryKey] ,StandardCost FROM[dbo].[DimProduct]
1 2 3 4 5 6 7 8
-- Rank Products based on the minimum and maximum cost per Product Subcategory SELECT[ProductSubcategoryKey] ,RANK()OVER(PARTITIONBY[ProductSubcategoryKey]ORDERBY StandardCost,[ProductKey])AS sequence_min ,RANK()OVER(PARTITIONBY[ProductSubcategoryKey]ORDERBY StandardCost DESC,[ProductKey]DESC)AS sequence_max ,[ProductKey] ,StandardCost FROM[dbo].[DimProduct] WHERE StandardCost ISNOTNULL
Averages excluding certain ranks:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
-- Get the Average and Qty. of the products per category excluding the minimum and maximum cost per Product Subcategory (above) WITH CTE_sequence AS( SELECT[ProductSubcategoryKey] ,RANK()OVER(PARTITIONBY[ProductSubcategoryKey]ORDERBY StandardCost,[ProductKey])AS sequence_min ,RANK()OVER(PARTITIONBY[ProductSubcategoryKey]ORDERBY StandardCost DESC,[ProductKey]DESC)AS sequence_max ,[ProductKey] ,StandardCost FROM[dbo].[DimProduct] WHERE StandardCost ISNOTNULL ) SELECT[ProductSubcategoryKey] ,COUNT([ProductKey])AS Quantity ,AVG(StandardCost)AS Average_Cost FROM CTE_sequence WHERE sequence_min >0 AND sequence_max >0 GROUPBY[ProductSubcategoryKey]
There are more functions that can be used and the user even has the option to dice the data further by using HAVING and slicing with GROUP BY CUBE | ROLLUP which plots the data in “pivotable” format.