Re: New IBM Nonsense

On 2005-01-23, DA Morgan <damorgan_at_x.washington.edu> wrote:
> thu.nnguyen_at_gmail.com wrote:
>
>> DA Morgan wrote:
>>
>>>thu.nnguyen_at_gmail.com wrote:
>>>
>>>
>>>> As for more
>>>>clients/spindle .. well if the system can handle it, they would
>>
>> need
>>
>>>>more clients/spindles.
>>>
>>>So your statement is that disk i/o is irrelevant to benchmark
>>
>> results.
>>
>>>Please point out the source of this fascinating revelation.
>>
>>
>> Of course not, but if you put in extra spindles or add extra clients on
>> a system that is already CPU saturated, you won't get any better
>> performance. You obviously don't know much about performance tuning.
>
> Of course I don't. So where, specifically, has anyone in this entire
> thread or an anything published by any company anywhere stated the
> condition you just created from your Ouija board that indicates CPU
> was saturated?

        It's reasonable to assume, given the sort of resources thrown at these little parlour tricks, that if Oracle could have thrown hardware at a benchmark to get it to perform better than it would have. They don't need little shills sticking up for them.

[deletia]

-- 
     If you think that an 80G disk can hold HUNDRENDS of           |||
hours of DV video then you obviously haven't used iMovie either.  / | \