![]() | |
#1
| |||
| |||
|
|
i am thinking about introducing historical data in an already existing database similar to what a version control system does for source code management. [...] before i start to reinvent the wheel i would like to ask if anybody can point me to some online references that describe solutions for this problem? |
#2
| |||
| |||
|
#3
| |||
| |||
|
|
before i start to reinvent the wheel i would like to ask if anybody can point me to some online references that describe solutions for this problem? |
#4
| |||
| |||
|
|
i am thinking about introducing historical data in an already existing database similar to what a version control system does for source code management. to be more specific: in the database there will be stored a given configuration of hardware elements and relations between these hardware elements. over time the hardware configuration will very likely change, e.g. old sensors will be replaced by new ones which have different callibration data and different identification codes. but i don't want to loose the old data and i would like to be able to ask questions like: what was the active configuration 12. march 2002? |
#5
| |||
| |||
|
|
In article <bg8l5k$5tf$1 (AT) pheidippides (DOT) axion.bt.co.uk>, programmer int.consultNOCAPITALS (AT) macmail (DOT) com> writes i am thinking about introducing historical data in an already existing database similar to what a version control system does for source code management. to be more specific: in the database there will be stored a given configuration of hardware elements and relations between these hardware elements. over time the hardware configuration will very likely change, e.g. old sensors will be replaced by new ones which have different callibration data and different identification codes. but i don't want to loose the old data and i would like to be able to ask questions like: what was the active configuration 12. march 2002? Isn't that what a data warehouse does? Absolutely NOT !!! Imagine a real-world database, shipping loads of goods. They have stock information, invoices, shipping manifests, and other assorted goodies. All this is transferred to a data warehouse, sliced, diced, and analysed ninety ways from Sunday, so that the Company Board can ask all sorts of useless questions and get a meaningless answer back. Okay, I'm being facetious, but the purpose of a data warehouse is to store oodles of pre-analysed data to enable quick answering of queries. Oh, by the way, some of the biggest are built on Pick technology, because the quickest way to bring a SQL db to its knees is to ask a question that the database hasn't been optimised to answer... Cheers, Wol |
![]() |
| Thread Tools | |
| Display Modes | |
| |