Re:How NoSQL forced the evolution of a scalable relational d...
by matwood on Friday July 13, 2018 @07:16AM
(#17521940)
> As a sidenote local cultural issue, the fact that we can operate the MongoDB servers ourselves, whereas RDBMS instances are with a central team and buried under a layer of bureaucracy was also probably an operational consideration.
This is clearly an oversight in the organization and will get fixed. I don't see it as neutral in the argument for or against some system.
> But where we experienced pain was when the business decided they wanted to do live reporting. We ended up piping the data into a SQL Server instance and using SSRS.
This is common even in RDBMs. You have a database designed to handle the running of whatever application. That design may not be the most efficient for reporting off that application. Flattening the data, and pre-aggregating is usually a good first step. Many data warehouse solutions can do some of this automagically if set up with a common schema type (like star).
Re:How NoSQL forced the evolution of a scalable relational d...
by lyqwyd on Friday July 13, 2018 @03:31AM
(#17521121)
The fact that you could operate the mongo instance, but not the DB is definitely a cultural issue, or more likely a managerial issue. Definitely NOT a DB issue. I’ve worked at places where we could manage the DB ourselves, and other places where we couldn’t eve talk to the DBAs without beuraucracy. It’s all about the management in those scenarios. And the company results were as you would expect.
When it comes to joins, these have lately been added to MongoDB, as has SQL - although the functionality is still rudimentary compared to a mature RDBMS.
But where we experienced pain was when the business decided they wanted to do live reporting. We ended up piping the data into a SQL Server instance and using SSRS.