When we launched MagnumBI our goal was to make Business Intelligence as simple as possible for our customers.
As a result we have ended up building a fairly complex architecture under the covers. Why you might ask?
If you think about Dropbox it’s a very simple product to use. It creates a folder on your laptop/desktop and anything you put in that folder is automatically sync’d to the Dropbox servers in the cloud.
And once its sync’d it’s then automatically sync’d to any other device that is subscribed to that folder, whenever that device is connected to the internet. All without the user of either device having to do anything. And once it’s sync’d you can also access it via any mobile device.
To do this they had to make sure syncing worked on any device, Windows, Windows mobile, Mac OSX, Mac IOS, Android etc.
They also had to make sure that syncing was robust and foolproof, if objects got lost or users lost trust that their data was safe then nobody would use the service. And this had to work regardless of the type of internet connect the user had, and whether they were permanently connected or not.
They had to make it automatic, so the user didn’t have to think or act to
Lots of complex technical issues to solve to make Dropbox seem like such a simple service to a user.
There is a great article that discusses the effort it took Dropbox to deliver this simplicity here.
For MagnumBI we had some complexity to solve to make it seem simple to our users. Things such as:
* loading data automatically from on-premise data sources
* provisioning new environments automatically
* scaling up or down at will
* integrating a number of 3rd party products
* redundancy everywhere to ensure high availability
* automated patching with minimal downtime
Ill blog later on the architecture we ended up creating to make Business Intelligence simple.