Here is more information about HFS+ than you'll ever want to know . . .
maybe:
http://developer.apple.com/mac/library/technotes/tn/tn1150.html#CoreConcepts
Brian
Tod Hopkins wrote:
> Brian was right, but I couldn't resist testing as I had the perfect
> opportunity.
>
> Empty 3GB RAID 0, inside brand new 8-core. Tried 16, 32, 64, and 256k
> stripes. Using AJA System Test set to 1080-8-bit, with file sizes
> from 512k-8GB
>
> At all file sizes, 32k was the winner or within 1% on read speeds.
>
> Overall, high to low numbers on all tests varied by only about 2%
> except for one or two outliers. In general, larger stripes favored
> larger files and smaller stripes favored smaller files.
>
> The difference writing files was more significant with the 32k size
> trailing at all file sizes by .5% on smallest to about 3% on larger
> sizes.
>
> The outlier was the 16k stripe which was significantly inferior for
> both 512k and 8GB files. It was competitive at 1 and 4GB. Go
> figure. The killer was that it was almost 5% slower reading at the
> 8GB file size and 2-3% at the 512K size.
>
> Inquiring minds and all. The unanswered question is block size. The
> block size in all cases is, of course, 4K, and that may well be the
> controlling factor.
>
> Cheers,
> tod
------------------------------------
Please donate to the Red Cross to help those in earthquake ravaged Chile: http://tinyurl.com/y8pgceq
Search the offical complete Avid-L archives at: http://archives.bengrosser.com/avid/
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/Avid-L2/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/Avid-L2/join
(Yahoo! ID required)
<*> To change settings via email:
Avid-L2-digest@yahoogroups.com
Avid-L2-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
Avid-L2-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
No comments:
Post a Comment