1. Home
  2. Relational Databases
  3. Microsoft SQL Server Reviews
  4. User Review of Microsoft SQL Server
SQL Server Serves Data Well
December 09, 2019

SQL Server Serves Data Well

Michelle Sollicito | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User

Overall Satisfaction with Microsoft SQL Server

We currently use SQL Server in Azure which makes management easier, though I have used on-premises SQL Server for many years, many versions. I am Microsoft certified in SQL Server 2012 and 2014. As a software developer I like the way SQL Server is easy to program against -- especially when using Entity framework or Entity Framework Core as the middle ware. As a database administrator I like the fact that SQL Server is truly relational, which makes it much easier to model business objects as data tables and to manage Referential integrity, locking and consistency. I like its ability to help with performance scaling via pre-compiled queries, stored procedures and triggers. I like its ability to help with scaling via sharing and replication. But above all I love SSMS (SQL Server Management System) which brings together all the tools in one place with easy access to query plan tools and analysis tools as well as SSIS / SSRS allowing integration of data with other systems and generation of reports.
It has matured over the years into a tool that makes life easy for those architecting or writing systems.
  • Modeling and Schema creation
  • Query optimization tools
  • Integration with coding tools via Entity Framework etc.
  • Analysis, integration and reporting tools
  • Better handling of blobs so that they do not slow down queries (lazy loading by default perhaps)
  • Automatic Migrations -- migrations are complex right now, tools that make it easier would be useful
  • Cheaper!!! It is too expensive, forcing many people to go to datalakes/s3/NOSQL databases when they do not really want to
  • There is no value you can put on data integrity. If data is invalid it can cause all kinds of problems in a business -- you lose customer trust, you lose market perception, you lose employee confidence. With data integrity and consistency, you eliminate all these problems but it's difficult to measure as an ROI.
  • Migration to and from SQL Server can be difficult if the original database was structured badly, but that is the case for any database.
  • Where tables were designed badly we have had performance problems, but that is because of the table design not due to SQL Server -- that would happen in any database. SQL Server does give you the tools to fully tune your database.
SQL Server is better for large databases containing structured relational data. It makes it easy to group and order, to sum and create tables of data from any data stored in a table or related tables. While Dynamodb is very good at STORING huge amounts of unstructured data, it is very difficult to use that data in any other way than as a group of related documents. It is very document centric. It is difficult to get totals and do queries by attributes (columns in a relational database). Firstly, because there is a need to do a map reduce first to find the documents to apply the query to and, then, because the attributes of the documents are often not easily accessible. Aurora, and to a lesser extent MySQL, are smaller versions of SQL Server to all intents and purposes, that run on any platform. They use the same query language.
So much support -- so many articles out there, so many sources, sample code, etc.

Do you think Microsoft SQL Server delivers good value for the price?

No

Are you happy with Microsoft SQL Server's feature set?

Yes

Did Microsoft SQL Server live up to sales and marketing promises?

Yes

Did implementation of Microsoft SQL Server go as expected?

Yes

Would you buy Microsoft SQL Server again?

Yes

Green field sites where you can start by modeling the business objects up front and getting the relationships right, the integrity right. Database First Entity Framework environments.

It's not so good for Code First or even Model First Entity Framework because typical coders do not really understand data modeling to the degree necessary -- and later on they suddenly find a whole load of duplicates they did not expect -- or a relationship that does not make sense, constraining what they wanted to do.