SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?


To: Tenchusatsu who wrote (62580)6/20/1999 11:45:00 PM
From: Charles R  Respond to of 1572159
 
Tenchusatsu,

<That's one advantage of a shared multiprocessor bus. So for two-to-four-way systems, Xeon will exhibit lower latencies than K7>

I would like to see some data on this before such a judgement can be passed. K7 has an additional flag, compared to P6, for its cahe coherency protocol to identify "shared owned" data (my recollection from old Microprocessor Report K7 article). This should go a long way in increasing K7 performance in multi-processor shared memory server apps. If it is a message passing architecture a la Chorus then cache coherency does not matter.

Chuck



To: Tenchusatsu who wrote (62580)6/20/1999 11:49:00 PM
From: Ali Chen  Read Replies (2) | Respond to of 1572159
 
Ten, < AMD's point-to-point bus and multiprocessor performance,>

Man, do you have something better to spend your
time instead generating irrelevant FUD?
Last resort on your sinking ship? (Just joking...)

<all the other processors must be notified separately>
What makes you think that the "north" bridge
can't send these messages simultaneously?
Remember, every processor has a separate "bus",
is not it?

<After doing a little research into the subject, I learned a little something interesting about the K7's point-to-point bus>
First, you seem to learn really "a little".
Second, how did you manage to do a research on
a non-disclosed-yet system architecture?

Please stop your groundless FUD.



To: Tenchusatsu who wrote (62580)6/21/1999 12:45:00 AM
From: kash johal  Read Replies (2) | Respond to of 1572159
 
Tench,

Re: extra latency for P2P.

Well an interesting proposition.

How do the TPC numbers for multiple Alpha's stck up?

I should imagine that would be a pretty good clue.

Regards,

Kash