Re: Huge Dataset (perhaps?) made for Range Partitioning (Long)

[JinNo Kim]

Edwin DeSouza wrote:
> JNK,
> For additional insight - can you describe the
> usage patterns for the Partitioned data:
>
> 1) In continuous CRUD use by transactional
> systems.
>
> 2) Once written - never updated. (i.e. Read
> Intensive).
> 2a) Used by Data Warehouse appilcations for
> Business Intelligence type work
> 2b) Used for Security Audits in case of security
> breach
> 2c) Used by SysAdmins to determine what caused a
> particular problem
>
> - Edwin

For this application, the data will be written once and basically will never be updated (except on extremely rare occurences) so "2" applies best.

Once we have a decent store of data, we hope to be able to identify trends and patterns that will help us make predictions about how to make other aspects of the business more responsive to recurring periodic fluctuations in demand, bringing 2a into play.

We'd also like the use the data to identify misappropriation of company resources when a routine audit turns up slight, but normally acceptable anomalies in the transactions from one of our locations. I guess this also more closely relates to 2a than b or c, but the audit aspect is there.

If you need more details about the specific application or how we intend to use it, I'd be happy to discuss it offline through email with you and the rest of the MySQL team, but would prefer to leave specific details off the forums. Send me an email at:

jinno_kim at yahoo dot com dot au

with the email addrs of those you'd like included in that discussion and I can provide more details through email. My boss has no issue with me discussing the problem in generalities here and I have every intention of sharing as much as possible within the public forum, but business is still business :).

Hope this provides a little more insight into what we intend to do with the database and how the data is used. I look forward to future communications,

-JNK