DragonFly users List (threaded) for 2005-02
Re: Dragonfly and Hyperthreading....
In a message dated 2/20/2005 11:25:26 PM Eastern Standard Time, Matthew
Dillon <dillon@xxxxxxxxxxxxxxxxxxxx> writes:
>:On Sun, Feb 20, 2005 at 12:35:22PM -0500, EM1897@xxxxxxx wrote:
>:> Does dfly, in its current state of development, yet take
>:> advantage of hyperthreading (single processor) within
>:> its kernel processes? My particular interest is networking
>:Nope. Don't expect any gains from HTT even for userland, it's generally
>:not worth it.
> I don't expect we will be doing any serious hyperthread-specific
> development. The cpu makers are clearly moving towards multi-core
> and the performance boost you get with hyperthreading is fairly small
> relative to what you get with multi-core.
Ah, but with hyperthreading the performance boost comes at zero
cost. The difference in cost between a 3.6Ghz system and "something
faster" is perhaps 50%.People pay $300 more for a 3.6Ghz xeon than
a 3.4Ghz. A small gain has significant value. And dual cores will be very
expensive for a long time. Intel and AMD aren't making them to
cannibalize their own business; they're doing it to have something to
make their margins on.
Matt, would you comment on the following article:
benchmarks are usually misleading, but at least its not some kid in his
basement (probably not) doing this one.