Replication is an amazing topology, it is one of the most used technology when it comes to making available data at another site or data center. I have had the opportunity to author a book on High Availability back in 2010, so in this post, I am not going to delve much into explaining it. In this post, let us understand, one of the errors that I have recently encountered during my weekend on-call. Data Validation. This very feature, operation in Transaction Replication is one of the features that help us determine whether the data being replicated to another end is really, exactly identical. Data Validation is done by either row count or checksum. Let us understand What is data validation in transaction replication.

What is data validation in transaction replication

What SQL Server Database Engine will calculate row count of both – Publisher and Subscriber.  This doesn’t include the data in text,ntext or image columns. There is another method – Checksum. Both methods – Row Count and Checksum match the columns or row count and then it compares it with the publication if it matches it is passed else it is failed.

Row Count

It is the very simple approach, it will count the row at publisher and subscription, compares it and pass or fail validation based on the count.

Checksum

The checksum is basically an algorithm that generates the algorithm values known as hash and compares it to the values original source of data and ensures that the data are genuine and error free.

I hope you have enjoyed reading this blog post so far. In the next post, next week we will understand how to fix data validation. Until then, I would encourage you to browse through HA&DR category of this blog space for few more articles on this very topic. And, do not forget to let me know if you like this blog post or not via comment section below?