Skip to content

Contact sales

By filling out this form and clicking submit, you acknowledge our privacy policy.

Can I shift my SQL language skills from on-prem to Azure?

The answer is yes! We cover what you’ll need to know. Also, we answer that pivotal question: What breed of dog would each Azure solution be?

Jun 08, 2023 • 10 Minute Read

Please set an alt value for this image...

The answer is yes! We cover exactly what you’ll need to know. Also, the answer to that pivotal question: What breed of dog would each Azure solution be?


Your keys to a better career

Get started with ACG today to transform your career with courses and real hands-on labs in AWS, Microsoft Azure, Google Cloud, and beyond.

If you’ve been a SQL Server pro for any length of time, you know that most of your job is dealing with bad news — either by responding to it or trying to prevent it. 

So, for once, let’s start with some good news! Those unfamiliar, intimidating cloud data technologies you have, so far, managed to avoid? They’re not as hard to use as you think. In fact, you can query the data in a lot of Azure data resources through some dialect of SQL. That means if you’re looking to make a change, you’re not starting from scratch.

Now, there is some bad news: If you’ve built your reputation on performance tuning, tweaking indexes, normalizing data, and traversing the secret tunnels and passageways of SQL Server, you may need to adjust your professional identity a bit. 

Not a whole lot, just a little.

Why? The platform as a service (PaaS) offerings in Azure abstract away a lot of complexity for relational databases. Auto-tuning and similar features are for real, these days; you’ll see Azure SQL is much less likely to shoot itself in the foot. For non-relational databases, many tweaks and optimizations simply don’t apply. 

(The term ‘non-relational’ is often synonymous with NoSQL, but the former is better. It’s a more accurate description of the data formats that differ from the highly structured, relational rows and columns of SQL Server tables.)

Despite my purported good news and only-sort-of bad news, you may have developed the wrong impression of cloud data resources. The likely cause? Dreaded employment ads that read like a vocabulary test, not a list of technologies.

The last time I looked at data-related job postings, there were requirements like:

"Candidates must have hands-on, professional experience in at least 13 of the following technologies: SSIS, SSRS, SSMS, SSAS, SSMinnow, TSQL, PL/SQL, ANSI SQL, noSQL, MySQL, yourSQL, ourSQL, weAllGotSQL, Mongo DB, HANA, Spark, Hadoop, Azure, Cloudera, MapR, Hortonworks, HortonHearsAWho, Informatica, Tableau, Crystal Reports, SAS, Python,  R (S, T, U, and V also helpful), Java, Ruby, Julia (AuntBea a plus), MATLAB, C/C++, Perl, Octave, Scala, GinsuKnife, F#, C#, B-flat."

Ok, maybe I’m being a little hyperbolic. My main point is that you really and truly do not need to know all, or even most of those languages and technologies to make the move to Azure. 

(I’m referring to the real ones. HortonHearsAWho is not a real technology; you knew that, right?) 

With only a little bit of training, you can readily begin reaping the benefits of data management, storage, and computing in the Cloud. 

Also, no matter what anyone else says, avoid qualitative assignments, such as “better” or “worse,” when it comes to data technologies. Stick with “most appropriate.” The fact is, in the 21st Century, SQL Server (on premise or in the cloud) is sometimes still appropriate, and sometimes it is not. 

Choice is a good thing!

There are no bad dogs

Consider the idea of assigning dog personalities to database technologies. Stay with me here!  I’m attempting to demystify some of the more popular Azure data offerings.

My first database experience was a very early version of FoxPro (you might need to use the Wayback Machine for this one). It was a lot like my family’s first dog, Muffin. Muffin looked a little like a fox, but more importantly, he was a basic suburban mutt. FoxPro was also pretty basic.

In the 70s and 80s, we fed our pets horrible, cheap, dog food or people food — and, well, it showed in their daily deposits in the backyard. GIGO at its worst, but that apparently preserved him for 17 years. Who am I to criticize? Muffin was a bit unruly if not kept on a leash and, like FoxPro, he was messy, leaving muddy pawprints and hair in his wake.

Fast forward a decade or so and consider Obie, the yellow lab. He was our SQL Server dog: reliable, loyal, trained, predictable. He ate higher quality, more expensive dog food and was a perfect model of not-GIGO. As long as we gave him love and attention, there was peace and harmony in the household. We could talk to Obie, ask him questions (one might say we queried him if one were prone to tortured metaphors), and we liked to believe he understood us. We also didn’t have to explain him to anyone. Everyone knows a yellow lab when you see one.

Then, my first Cloud data experience was blob storage. Unstructured, big data, stored securely and simply. Not unlike our next dog, Gruner, a beautiful Great Pyrenees, who spent most of his time sleeping.

But when the goats or the chickens needed protection (or the doorbell rang), he literally sprang into action. He had some admirable features and he could run like the wind. Like the images kept in a blob storage container, Great Pyrenees are a visual medium. Body language was more important than verbal communication.

All of them were unique, none of them necessarily better than each other. 

I won’t bore you with additional pictures of our family dogs, but here’s a quick summary of a few of the more popular data resources in Azure. It’s not an exhaustive list, but should give you an idea of some of the skills and languages you need to make them a part of your cloud or hybrid solutions. 

As an SQL Server professional, you’ll find you really are not that far off the mark in terms of training and experience.

The Labrador Retriever: SQL Server on a Virtual Machine

Pet Personality: Predictable, reliable, and coachable.

Type and Purpose: A relational database optimally used for transactional data, where strong consistency and enforced structure is important. Because installation on a VM makes this an IaaS (infrastructure as a service) offering, this SQL Server option may be best for customers who want to retain more control over the maintenance, tuning, and typical on-premise database administrator duties. 

Alternative To: Azure SQL, Azure Database for MySQL, for PostgreSQL, for Maria DB

SQL (or SQL-like) Support: Yes

Product Information: Click here

The Labradoodle: Azure SQL Family

Pet Personality: Similar to the Labrador Retriever, but smaller, with less fur to vacuum.

Type and Purpose: Relational databases are optimally used for transactional data, where strong consistency and enforced structure are important. As a PaaS offering, with a fully “serverless” option, the platform is fully managed. However, some control is retained by customers who choose the Managed Instance sku. 

Alternative To: SQL Server, Azure Database for MySQL, for PostgreSQL, for Maria DB

SQL (or SQL-like) Support: Yes

Product Information: Click here

The Golden Retriever: Azure Database for MySQL/PostgreSQL/Maria DB

Pet Personality: Disposition and qualities similar to a Lab.

Type and Purpose: Relational databases are optimally used for transactional data, where strong consistency and enforced structure are important. Fully managed or mostly managed, depending on the sku. Ideal for quick migration of existing databases already under one of the three technologies.

Alternative To: SQL Server, Azure SQL

SQL (or SQL-like) Support: Yes

Product Information: MySQL, PostgreSQL, MariaDB

The Bernese Mountain Dog: Azure Synapse Analytics

Pet Personality: Massive, hard-working, versatile dogs for farms in native Switzerland. They excel at drafting and carting.

Type and Purpose: A MPP (massively parallel processing) analytics platform with the storage component based on SQL Data Warehouse and a bundle of other services, such as Azure Data Factory. Well suited for analytics workloads over both relational and non-relational database resources.

Alternative To: Azure HDInsight, based on an Apache open-source platform which is also an MPP service but with fewer bundled services. Apache technologies lead in many big data applications, and this is the source of many of the languages and buzzwords that may be unfamiliar to SQL Server professionals. But don’t let that stop you from getting started in Azure. You’ll learn what you need to know when the time comes.

SQL (or SQL-like) Support: Yes

Product Information: Click here

The Jack Russell Terrier: Azure Data Factory

Pet Personality: One of the best agility dogs on the obstacle course.

Type and Purpose: A highly flexible, fully managed data integration tool used for both traditional and big data ETL workloads.

Alternative To: SSIS, which can be hosted inside of an ADF pipeline.

SQL (or SQL-like) Support: Not necessarily applicable to all cases, but supported where it makes sense. If you are familiar with SSIS or other “workflow” style data integration tools, you can pick up ADF, with some training.

Product Information: Click here

The German Pinscher: Cosmos DB

Pet Personality:  The “Swiss Army Knife” of working dogs. A breed that loves to work and is adept at many different jobs.

Type and Purpose: A multi-model database, with unlimited elastic scale and global distribution. Suitable for a wide variety of use cases, from gaming to eCommerce and transactional workloads. “Multi-model” because the native format is JSON, with SQL querying, but also has other APIs for Cassandra wide-column, Gremlin graph, PostgreSQL, Mongo DB, and key-value pair formats.

Alternative To: Other document databases, but nothing similar in Azure.

SQL (or SQL-like) Support: Yes, the Core API is SQL over JSON documents.

Product Information: Click here

The Great Pyrenees: Azure Blob Storage and Azure Data Lake Storage Gen2

Pet Personality: May sleep a lot, depending on job assignment, but quick on the switch. Has a sprinter’s physique and a marathoner’s endurance.

Type and Purpose:  Blob storage is optimized for storing, serving, and streaming massive amounts of unstructured data, such as audio, video, images, and log data.

Alternative To: Nothing similar in Azure.

SQL (or SQL-like) Support: Yes, sort of, but let’s not go there. There are better ways to work with unstructured data. This will be a fun training opportunity!

How you can take the lead

So, how do you brush up on those small gaps that might still exist in your knowledge? I’d highly recommend checking out A Cloud Guru’s library of Azure courses and hands-on labs. There’s learning resources for wherever you are in your Azure adventure.

If you’re interested in certifications (and who doesn’t love another bit of paper telling them they’re great?) I’d suggest reading this wonderful article by Lars Klint: “What’s the Best Microsoft Azure Certification Path For Me?

Want to keep up with all things Azure? Subscribe to A Cloud Guru on YouTube for weekly Amazon news and AWS announcements. You can also like ACG on Facebook, follow us on Twitter, or join the conversation on Discord!

Amy Coughlin

Amy C.

Amy Coughlin is a Pluralsight Author and Senior Azure Training Architect with over 30 years of experience in the tech industry, mainly focused on Microsoft stack services and databases. She's living the dream of combining her love of technology with her passion for teaching others.

More about this author