To my knowledge there are these 2 ways you can go:
If you want performance, simpler queries, easier programming, then you should make a second table with your ID as foreign key and make there a column for each of your properties that you want to store. For any new attribute you would have to change the database schema, what might be a major drawback.
The above won't work, if you need flexibility in your database schema (because your meta properties are not known in advance or are point to change), but it will add to your code and queries complexity (depending on your framework/application) and also lead to poor performance compared to the first alternative. Poor performance can be compensated to a certain extent with more hardware, but that's expensive.
It's a trade off between the two. You must judge your application scenario and decide what you need most and what is a absolute no-go. Then decide on the remaining criteria what you want to use.
I have seen and used both ways I mentioned above, and must say the second one worked better for me (didn't have high volume querying though), even if it is a antipattern from a SQL / database normalization point of view. However, we aren't payed for nice SQL database designs, but for working applications.
I think the difficultly lies in using the parent table as part of the sub-table's primary key. So long as you have a reference to the parent table, you can always generate a reference that combines Location+Room or Room+Rack. Otherwise the primary key will get too hard to update and you have bad primary key.
I can see two ways to tackle this depending on how many locations, rooms, racks you have. It depends on quantity for your interface.
Firstly, a schema...
tblLocation(locationID[PK++], locationReference)
tblRoom(roomID[PK++], keyLocation[tblLocation.locationID], roomNumber)
tblRack(rackID[PK++], keyRoom[tblRoom.roomID], rackReference)
tblDevice(deviceID[pk++], keyRack[tblRack.rackID], positionU, deviceLabel)
...tables for custom devices can extend tblDevice
Foreign keys in square brackets, PK++ means an auto increment primary key.
As you can see, a Location has many Rooms, a Room has many Racks, a Rack has many Devices.
I have made primary keys as auto IDs as I prefer them to be integers and there's less to think about should you change a location/room/rack reference.
Populate this with Locations, Rooms and Racks as appropriate.
Solution 1 :: Not a lot of racks
- Have three drop downs on the form.
- User chooses a location, you then select all the rooms of that location in the next combobox.
- User then selects the room, you then fill the next combobox with the racks in the room.
- User then fills in device details and U position (which you probably want to validate to make sure that you don't have anything allocated there.)
Solution 2 :: Lots of racks
Have a query similar to:
SELECT
rackID,
locationReference + '-' + roomReference + '-' + rackReference AS [Full Location]
FROM tblLocation
INNER JOIN tblRoom
ON tblLocation.locationID = tblRoom.keyLocation
INNER JOIN tblRack
ON tblRoom.roomID = tblRack.keyRoom
ORDER BY
locationReference ASC,
roomNumber ASC,
rackReference ASC
To generate a list of Location-Room-Rack references.
ABC-101-A/1
ABC-101-A/1
ABC-101-A/2
ABC-101-A/2
ABC-105-B8
ABC-105-B8
FAY-208-X8
FAY-210-A9T
FAY-210-A9T
FAY-210-F4
NAT-410-78
NAT-410-78
User selects one, then fills in the information. List will be equal to the number of racks (which could get a bit excessive).
Advantages
- If you move a rack (or racks) of equipment from one room to another, it's just a simple update query to update the database
- If a location/room/rack reference changes, you only have one piece of data to update.
Further schema additions
I've used racks in the past, I simplified the schema but these are some additions that spring to mind.
- Add rackSize to tblRack: the size in U's
- Add sizeU to tblDevice: the size of the device in U's
You can then do further validation: if you have a 39U rack and somebody tries to put a 5U device in at location 38, you can throw an error (as U's 40-42 don't exist!). Also if somebody puts a 3U device in at location 10, and something is in location 11, you know it won't fit or they have not updated the details of the device that is no longer there!
You could also make some clever queries that would tell you how much space there is in your racks (and where). But that's beyond the scope of this question!
Hope this helps.
Best Answer
It may seem unwieldy, but I would go with the approach of having a separate table rather than using a comma-separated string.
In addition to avoiding the parsing complexities, this allows you to do nice things like apply foreign key constraints and / or indexes to that
user_id
column. It also simplifies mass updates if they are ever needed ("remove user 23 from all alerts").I would differ from your proposed schema in that I would leave the two boolean fields on the main alerts table, as they apply whether or not there are individual sets of recipients.