[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: Multithreading question
From: |
Darold Higa |
Subject: |
RE: Multithreading question |
Date: |
Wed, 29 Jan 2003 09:55:22 -0800 |
Just for the record Apple announced a price drop yesterday, and in fact all of
their non-server desktops are already multiprocessor.
I usually stick to AMD processors for price:performance, but some friends at
Intel had their hands on the 3.06GHz chips prior to public release and were
pretty impressed (I think these are the first to fully implement HT). That is
why I'm investigating my options. I understand that Intel's HT will give some
boost in performance even if your application is a monolithic block since OS
tasks are partitioned off to one of the virtual processors.
I don't trust Dell, I usually build my own. They tend to take shortcuts where
I don't want them to.
Thanks for the notice!
Darold Higa
-----Original Message-----
From: address@hidden
[mailto:address@hidden Behalf Of Christopher Mackie
Sent: Wednesday, January 29, 2003 8:21 AM
To: address@hidden
Subject: RE: Multithreading question
>> I vaguely recall parallel processing as one possible direction for
>> swarm.
>>
>> It seems that the two leading contenders hardware wise are Apple
>> desktops and a P4 system with Hyper Threading, both are multiprocessor setups
>> (albeit virtual in the P4 case).
Just for the record, I'm not aware that Apple is shipping multi-processor
systems at no upcharge, and there are certainly Intel multi-processor systems
available (I'm writing this on one). You can buy them (mine's a Dell), or
build them (check motherboards.com for multi-processor boards and chip
suggestions). They're cheaper than the Apple multi-proc solution, too,
although that dual-processor XServer really is a beautiful machine.
What I've seen about hyperthreading suggests that you'll probably get some
performance boost without special libraries; how much is never generalizable in
these cases (hyperthreading is optimized to accelerate Windows API threads, not
necessarily 3rd party apps). Whether you can get more boost with a
hyperthread-aware library I dunno, but I would certainly expect so.
Further, not all P4s have hyperthreading: you need the Northwood core
(Northland? I never can remember). Anything above 2.3GHz is Northwood, if I
recall correctly, but below 2.3GHz, machines may be either Northwood or its
predecessor. Don't rely on my memory: Intel's site covers the details and how
to tell.
I believe the latest generation Xeon processors (still PIII but optimized for
serving) also have hyperthreading, but I'm not sure: check the Intel website.
Next-gen Xeon processors, which will have hyperthreading, are due this quarter,
by the way, so you might want to delay a purchase 60 days or so.
And just to round out the chip review, AMD is supposed to release its 64-bit
desktop chip (32-bit apps still run native) sometime this summer. It's going
to be a lot more affordable than the Itanium, and the backward compatibility
will also be a plus, so those of us who need "long longs" and 64-bit
addressability are looking forward to it very much. I don't know about
hyperthreading with the AMD chip, but I suspect not.
HTH, --Chris
==================================
Swarm-Support is for discussion of the technical details of the day
to day usage of Swarm. For list administration needs (esp.
[un]subscribing), please send a message to <address@hidden>
with "help" in the body of the message.
==================================
Swarm-Support is for discussion of the technical details of the day
to day usage of Swarm. For list administration needs (esp.
[un]subscribing), please send a message to <address@hidden>
with "help" in the body of the message.
Re: Multithreading question, Marcus G. Daniels, 2003/01/29
- Ok, what about benchmarks, Darold Higa, 2003/01/29
- Re: Ok, what about benchmarks, Marcus G. Daniels, 2003/01/29
- RE: Ok, what about benchmarks, Darold Higa, 2003/01/29
- RE: Ok, what about benchmarks, gepr, 2003/01/29
- Re: Ok, what about benchmarks, Marcus G. Daniels, 2003/01/29