SQL Server has been one of my favorite software products of all time. As an admin it has been rock solid and not produced any headaches that weren't our own fault.
With that said, to really get the most out of SQL Server (and most likely most other SQL implementations) you really have to do your homework and put in the time to go through the features. Perhaps more than anything this is why NoSQL has taken off-- it's very simple to get going. At user groups I hear a lot of people saying they selected MongoDB because "They had millions of rows and SQL just couldn't keep up" and to me it just sounded like no one in their organization had any solid SQL experience.
It's too bad MS didn't build a 'SQL Admin' into their product that sent an email on occasion to say things like "You have a query that is called frequently and could be sped up if you simply included this column in this index. Here are some details!"
SQL Server has been one of my favorite software products of all time.
I feel the same way about PostgreSQL, along with the bit about learning all the features. I suspect that most decent RDBMS's are this way.
At user groups I hear a lot of people saying they selected MongoDB because "They had millions of rows and SQL just couldn't keep up" and to me it just sounded like no one in their organization had any solid SQL experience.
hahahaha. I have customers with db's with 10's of millions of rows (And I expect hundreds of millions or rows in the foreseeable future) and I can't imagine MongoDB keeping up in terms of reporting......
I feel the same about Oracle. There is just so much capability in the Oracle RDMBS if you take the time to learn it. If your organization is paying for SQL Server or Oracle and then just doing simple SELECT, INSERT, UPDATE, DELETE or throwing an ORM on it you are leaving a serious amount of value unutilized.
"At user groups I hear a lot of people saying they selected MongoDB because "They had millions of rows and SQL just couldn't keep up""
hehe :)
At work we do 500 million + in Mysql, billion(s?) if you count all the tables. But to be fair, you do loose a lot of flexibility and "features" when the tables become this big.
With that said, to really get the most out of SQL Server (and most likely most other SQL implementations) you really have to do your homework and put in the time to go through the features. Perhaps more than anything this is why NoSQL has taken off-- it's very simple to get going. At user groups I hear a lot of people saying they selected MongoDB because "They had millions of rows and SQL just couldn't keep up" and to me it just sounded like no one in their organization had any solid SQL experience.
It's too bad MS didn't build a 'SQL Admin' into their product that sent an email on occasion to say things like "You have a query that is called frequently and could be sped up if you simply included this column in this index. Here are some details!"