| Help with database holding a large amount of data [message #447612] |
Tue, 16 March 2010 08:55  |
clancypc
Messages: 36
Registered: December 2006
|
Member |
|
|
I am designing a database that will hold a large amount of data. The data to be held will be telephone call records. There will be approximately 112 million records per day and they want to store 90 days worth of data (so approximately 12 billion records).
The table design is pretty much set in stone, the business has told me which fields in the record it wants to store so there is pretty much just one table with all the fields that they want. There is not really anything to normalise.
Now I was planning on storing the data in partitions based on the start date of the calls. Then when the 90 days are up I can just drop the oldest partition.
So what I am worried or wondering about is the following:
Can I divide the data into partitions based on the date? (I think it can be done, just want to check)
Can a partition hold 112 million records?
Can a table hold 12 billion records?
What would be the best way to load 112 million records per day?
Can anyone provide some advice?
|
|
|
|
|
|
| Re: Help with database holding a large amount of data [message #447685 is a reply to message #447616] |
Wed, 17 March 2010 06:25  |
clancypc
Messages: 36
Registered: December 2006
|
Member |
|
|
|
Wow, that looks really useful. Unfortunately there is no guarantee that I will be getting Oracle 11 to work with. I currently have Oracle 10, so I may have to do it manually. I can automate it of course create the next days partitions on the current day at the same time as dropping the oldest partitions.
|
|
|
|
] 
Blog
Search
Help
Members
Register
Login
Home
