Re: Synchronous writes & TEMP

Frank,

[VC]
> >I'll give you a hint: Oracle can use both
> > asynchronous IO _and_ synchronous writes at the same
> > time when performing IO to the TEMP tablespace.

[FB]
> So? I mixed up delayed writes, synchronous writes, async IO.
> Why did you not use delayed writes vs synchronous writes in
> the first place? Would have made it clearer for morons like me.

One would imagine that the terminlogy's been sort of obvious for anybody familiar with OS basics at the undergraduate level but I guess I was too optimistic.

[FB]
> So my best guess is: it in the habit of Oracle to use
> synchronous writes.

I do not understand what 'habit' means in this context -- we are not talking about some creatures are we ?

[FB]
>
> On the other hand - would you like your benefits calculated
> wrong, because Oracle missed dome data in performing the
> group by statement (using delayed writes)?

Describe how it would be possible for Oracle to 'miss some data' with delayed writes to TEMP, the whole hypothetical chain of events please.

> Does it make any difference at all? You state it does, but
> we had this raw IO vs buffered IO thread not too long ago.
> I'd like to see some test results, presumably from a C-program
> (as Oracle will always use O_DSYNC)

Again, you are disputing something relly basic. Please read about why operating systems implement buffered I/O at all and you would not need to perform any such tests ("Advanced Unix Programming", by R. Stevens)..

The thread you are referring to is irrelevant since TEMP datafiles are different form others because:
a. one does not care about potential data loss (presumably, unless someone can explain to me where I am wrong);
b. I/O to those files is not cached by Oracle;

I do apologize for my harsh words but I was getting really frustrated with people's inablity to understand what I was trying to ask despite my failed attempts to clarify the meaning several times ...

P.S.:

[quote="Meriam-Webster"]

Main Entry: Em·bar·ras
Variant(s): or Em·bar·rass /'am-"bro--sic/ [/quote]

[quote="The Free On-line Dictionary of Computing, © 1993-2003 Denis Howe"]

orthogonal

adj. [from mathematics] Mutually independent; well separated; sometimes, irrelevant to. Used in a generalization of its mathematical meaning to describe sets of primitives or capabilities that, like a vector basis in geometry, span the entire `capability space' of the system and are in some sense non-overlapping or mutually independent.

[/quote]

Regards.

VC Received on Sat May 15 2004 - 07:55:24 CDT