Database Architecture for time-data

[IWeb]

Hello,
I want to store data from a heart-frequency-measuring device. This data is approx. between 1000-10000 rows per dataset. I get at least 100 datasets per day which results into up to 1.000.000 rows/day. I do not need to perform database related procedures onto this data - I just want to store and load each dataset.
Now I'm wondering, what's the best way to store this data. I have summarized the following methods till now:
- store each row as a sql row -> 1.000.000 rows/day
- store each dataset as a blob -> only one row per dataset in the database
- create for each user a table and store there each dataset row as a sql row.
- store each dataset as a file.

What method would you suggest? Are there better solutions?
Thanks,

[pete_bisby]

If you're looking at a way of storing information that can be made readily available and loaded/unloaded quickly, without the overhead of storing it within a database, I would choose storing the data as XML files. This will allow you to store the information in a uniform manner, and readily availble for searching and loading/unloading to and from database tables.

If the XML files have an associated DTD and XSLT, they can be displayed very easily in a web browser.

[Sxooter]

Well, if all you're doing is storing it, you could copy it to /dev/null and be done. After all, if you're not doing anything with it afterwards, it doesn't matter how you store it really.

So, there must be something you ARE going to do with the data, and if we know that we can make more informed suggestions on how you should store it. 🙂