Share via

What is the difference between Azure Purview and MDM \DQS in SQL server 2019?

Mudassar A 491 Reputation points
Mar 30, 2021, 3:05 AM

What is the difference between Azure Purview and MDM \DQS in SQL server 2019?

I see Azure Purview was released recently and is cloud DG\DC tool service.

Now if Azure Purview is the future of Datagovernance in Microsoft then what happens to MDM\DQS in SQL SERVER 2019?

Azure Data Catalog
Azure Data Catalog
An Azure service that serves as a system of registration and system of discovery for enterprise data assets.
106 questions
Microsoft Purview
Microsoft Purview
A Microsoft data governance service that helps manage and govern on-premises, multicloud, and software-as-a-service data. Previously known as Azure Purview.
1,300 questions
0 comments No comments
{count} votes

Accepted answer
  1. Erland Sommarskog 114.4K Reputation points MVP
    Mar 30, 2021, 9:54 PM

    As I understand it, Purview is a lot wider than MDS. Purview can scan your data estate, not only SQL Server but a few more data sources. It then catalogs your data, so you can find out what you have that is sensitive in various ways etc.

    I don't know much about MDS, but as I understand MDS is something you load data into yourself, and then you can use MDS as a common source of information in your organization, and it still has a place in this world.

    DQS was a complete failure from the start, but apart from that there is not much overlap with Purview.

    0 comments No comments

2 additional answers

Sort by: Most helpful
  1. Nick McDermaid 1 Reputation point
    Nov 22, 2021, 12:51 AM

    MDS - master data services lets you manage datasets and add reference information to that data. For example your source system might have a list of 2 million stock items, but the items stored in the system are missing the weight or useful categorisation.

    MDS lets you import these records and add additional data such as reporting categories and manage this in excel. Then you can use this for downstream reporting purposes. Or you can create completely new datasets that are required for reporting. AFAIK Purview doesn't do this at all, but there are lots of promises that it will become the platform for it, and it makes sense that it would be.

    Microsoft SQL Server MDS uses rather old technology and I would not recommend it for new work.

    Microsoft SQL Server DQS is even older and again I would not recommend it for new work. It seems like something they deployed to tick a box in Gartner ten years ago.

    Purview's current functionality appears to be limited to data profiling and tagging. What tables and columns are in your databases? what's the data profile? OK now add some tags so end users can easily search and find these columns.

    AFAIK Purview does not store any actual data (i.e. a list of stock items) except for limited row previews. It stores data about metadata. (what columns are in this table? how many rows does this table have?)


  2. Ricardo Silva 6 Reputation points
    Jan 17, 2024, 6:48 PM

    MDS enables sharing common information across different systems. It can be the source of truth or it can work as a hub, capturing inforrmation from one system and sharing to others. Also, it provides access control, versioning, workflows, and revisioning before publishing new content.

    DQS provides cleasing, profiling in a easy way to manage your data that currently Purview is not capable at all (Azure Data Factory introduced a very limited funcionality also not funcional without doing a lot of coding).
    There are 3rd party solutions in the market however they are extremely expensive while Purview is far from be able to replace them.

    Purview, as of today, provides some assistance in data governance by storing and managing metadata, data lineage and access control but it stops there.

    MDS and DQS are more important today than ever given the increasing amount of information with bad data quality that is collected daily by companies with issues with their data governance. They are excelent tools to integrate data contained in silos inside businesses while providing clean and uniform data across the board and reinforcing data ownership.

    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.