6 posts • joined 2 Jan 2007
Guys... You know that there's more than 1sq/in on a drive. You know that the stated (new density) is 1TB/sqin. So don't tell me that the capacity will be 16GB.
Also. Brett. Mate. I think you just multiplied the current density, by the area, to get current capacity. I like your thinking though. So lets try again
density * area = capacity
1 (TB per sq in) * 3.8 (sq in) = 3.8 TB. Better I think.
Now - for a multi-platter, multi side arrangement, we might expect four times that.
What value does this provide?
It's not humour... There's no new information.
I'm surprised also that he's not discovered SYSTIMESTAMP - which has been available for a while (and is more accurate than SYSDATE).
But yeah... There's nothing in this article except promotion of somebody, who promotes something else.
DBA and the model
One thing that may be forgotten here is that many DBAs in smaller businesses support off-the-shelf (albeit customized) software for much of the time. SAP, Siebel and the likes.
They probably haven't done many production enhancements in terms of actual code additions or reprogramming. For sure they patch systems. But more the database side than the application code. And ultimately, it's the app maintenance that a model (or lack of) affects.
I guess... at the end of the day... most businesses don't care. And to some extent they have good reason. For performance to matter tomorrow, we have to first meet todays goals. It's fair to expect that tidying up is left to stage two or three. It's up to us as developers to ensure that we isolate the app from the structure enough that a tidy up will actually be feasible later.
Not this old stuff....
Logical vs physical... what a title. This is a debate out of 70s.
Obviously - we should have the models. In truth, we probably don't have the models and no longer live in a simple OLTP/ER world. Instead there's XML, ROLAP, and ORDBMSs.
So much for that. We'll do models if we have time.
Now about that first post. What is this about DBAs doing the models. That's crazy. The developers should do the models. All of them. Perhaps with some DBA input but that's it.
It depends on the developers.
The answer is - yes, and no.
The features and flaws of each database product are truly significant in my view. But the vendors have another problem.
The average developer knows very little about the features. It has got to a point where existing functions can be rebranded and re-released to the developer community - somewhat in the manner of old Disney films.
Indexing is the only way to speed things up? Rubbish.
Try aggregates. Try partitioning - horizontal and/or vertical.
If that fails, sack your consulting staff and re-engineer your system with a small bunch of competent techs.
- Mounties always get their man: Heartbleed 'hacker', 19, CUFFED
- Analysis Oh no, Joe: WinPhone users already griping over 8.1 mega-update
- AMD demos 'Berlin' Opteron, world's first heterogeneous system architecture server chip
- Leaked pics show EMBIGGENED iPhone 6 screen
- OK, we get the message, Microsoft: Windows Defender splats 1000s of WinXP, Server 2k3 PCs