I’ve been thinking a lot lately about the incorporation of JSON into SQL Server. At first I see it as a technology to be abused. I can’t tell you how many stored procedures I have seen that had a single input parameter as an XML variable when XML was introduced to SQL Server. It was very popular in some circles […]

When space is an issue, something that doesn’t happen very often, we can optimize storage by taking advantage of bits instead of using an entire byte to store a value. In fact, depending on the number of values needed for a specific tuple, you can break up a byte into multiple values. If you had 8 bool values they could […]

I’m working on an application that uses a Dot Net FlagsAttribute class implementation for enumerated values. It works like this. You create an Enumeration using values with details having values assigned as sequential powers of 2. Enum colors { Black = 1, Blue = 2, Green = 4, } With this enumeration each item is unique. The difference is that […]

SQL Server can optimize the logic of your queries resulting in a better execution plan which you may not have considered. I wanted to prove that the AND operation in TSQL is the equivalent of the && operation found in C#. When you compare two Boolean values to determine if both are true you combine the two values using &&. […]

There is a very clear and present limit to how much performance you can achieve in an SQL Server simply by adding RAM. If you don’t have the necessary indexes or statistics on a table, the query plan produced by the Query Optimizer may be such that no amount of RAM will solve the problem. I found this to be […]

Can’t I just add a bunch of memory to my SQL Server instance and have the same performance as using in memory tables? It turns out that the answer is no. Adding memory, if your database can use it, will often help performance. Everything SQL Server does to a database file is created initially in memory, and then written to […]

One of the things I dislike about POCOs (Plain Old CLR Objects) is the limited encapsulation available in Dot Net. POCO classes tend to be quite anemic, often implemented using properties only. That’s because the purpose of a POCO class is to transport data in object form. Including other behaviors makes the POCO more powerful, but increases the complexity for […]

What problem is being solved by doing a migration first design? Eilenblogger asked this question in another way, “What do you do if your user acceptance tests fail? Do your migratons roll back, or do you simply restore from a backup?” This question is the crux of why I like to do database migrations first. I worked on a project […]

Today I want to introduce what I think may be a new term…Database Migration First Development. Most database developers use a number of different techniques to make changes to the next version of a database. This includes the very first instance of a database, as well as making modifications to an existing database. Here are some processes used in the […]

Free virtual conference….are you registered yet? We’ve been working hard to make sure you have the best sessions, the most up to date information and … yes, no travel required. No rubber chicken meals, no nights away from your work, no lost billable time. Watch, learn, interact, all new, all excellent sessions. It’s the SSWUG.ORG Fall Virtual Conference! When you […]