At 8G ram you are basically giving each client a max of 256M workspace. That might be a bit tight .... it depends on the application you are running really and how much of that application is per instance data and how much shared memory and such. However 16G ram would probably not hurt, or 2 systems with 8G ... too little ram also means the OS wont be able to cache enough IO which can be very bad in a thin client setup.
For disk the same principles are true. Solid state disks are good up to a certain level, but a raid mirror with a few fast sas disks is probably better if you are to do anything involving lots disk writes.
On Oct 10, 2010, at 11:56 AM, Daniel Howard wrote:
Bart, you hit the main issue, and why I started looking at the 6
cores in the first place. I'm sure that there is a CPU
speed/architecture difference where a 6 core CPU with lower speed
and less cache, etc. will perform worse than a 4 core at higher
speed, but these two options were so close in all but number of
cores and architecture that I wanted to make sure:
AMD Phenom II X6 1090T Thuban 3.2GHz Six-Core Processor, L2 Cache: 6
x 512KB, L3 Cache: 6MB, ($266)
Intel Core i7-950 Bloomfield 3.06GHz Quad-Core Processor, L2 Cache:
4 x 256KB, L3 Cache: 8MB ($295)
I'm planning on 8GB RAM, should be more than enough. For high disk
bandwidth, I'm still not clear on what would be best, especially
with the new solid state drives, any additional thoughts there
regarding SATA-II, SCSI, and Solid State? Best, Daniel
On 10/10/2010 11:38 AM, Bart Lauwers wrote:
Daniel,
Benchmarks represent typical desktop workloads. This means
they fail to leverage extra cores to full benefit because
desktop applications are (usually) not designed to take
advantage of multiple cores.
In your case, for a thin client server: a six-core cpu will
generally outperform a 4-core cpu since more tasks will be
trying to run at the same time.
A 4-core 1Ghz cpu will beat a 2-core 4Ghz cpu any day when it
comes to workload. But in benchmarks the 2-core will always win.
With 32 clients this will be very pronounced as you will have
50% more jobs that can run at the same time on a 6-core vs
4-core. The tiny bit of speed gained on the intel would not
match the extra overhead of 50% more workload switching.
Finally, you want to make sure you have enough ram and disk
bandwidth. If you underbuy disk or ram, the cpu won't matter.
And yes, I've implemented this in practice.
Bart.
On Oct 10, 2010, at 9:10 AM, Daniel Howard wrote:
Thanks Tim, that
confirms what my gut felt, that even though 6 cores from AMD
is more, architecturally the 4 cores from Intel are a better
performer. But it really depends on how the software uses
the multiple cores, so that's why I'm hoping someone may
have actually tried it.
Alternately, if someone has used a single quad core
effectively for serving 32 simultaneous clients in a lab
setting, then it's just a cost thing.
Best, Daniel
On 10/10/2010 3:01 AM, j. Tim Denny wrote:
Daniel
Someone did all the work for us...
seems the 4 core Intel is a better deal contrary to
the price/benchmarks I previously posted.... on that
page they show a $199 quad core intel beating a hexa
core AMD at $285
T
__________________________________
j. Tim Denny, Ph.D.
Consultant -
International Development, Education and ICT
SKYPE - jtdenny Googletalk - denny.jt
http://www.linkedin.com/in/jtdenny
https://www.avuedigitalservices.com/VR/id130765695
.....
"People
generally quarrel because they cannot argue."
Gilbert
Chesterton (1874-1936) English writer
On Sun, Oct 10, 2010 at 2:24
PM, j. Tim Denny <johndenny@xxxxxxxxxxxxxx>
wrote:
Daniel
very interesting question as it brings up so
many issues.... you are asking to compare two
CPUs for the same purpose and then want to know
which one is more appropriate..
I was just browsing CPU benchmarks...not
knowing which chips you are comparing I can only
take a guess...
here is what I see...
So if
benchmarks mean anything then the AMD listed is
25% faster and 20% cheaper.... but what about
power draw and any other issues that may come to
play....
you
then ask how about LTSP performance... I
wonder... is the OS optimized for multicore
usage? or does that matter?
But
then what about GPU? are thin client
environments dependent on GPU performance or
does that not matter cause each thin client has
it own mainboard?
Tim
__________________________________
j. Tim Denny, Ph.D.
Consultant -
International Development, Education and ICT
SKYPE - jtdenny Googletalk - denny.jt
http://www.linkedin.com/in/jtdenny
https://www.avuedigitalservices.com/VR/id130765695
.....
"People generally
quarrel because they cannot argue."
Gilbert Chesterton
(1874-1936) English writer
On Sun, Oct 10, 2010 at
10:04 AM, Daniel Howard <dhhoward@xxxxxxxxxxx>
wrote:
Anyone have any experience comparing
performance of a 6 core AMD CPU-based thin
client server to an Intel quad core based
server? I'm looking for a new 32-client
computer lab server. I lean to the quad
core since it's at least a generation ahead
architecturally, but wonder if 6 cores gives
better LTSP performance when students are
really just doing OpenOffice and
Firefox/Chrome 99.9% of the time. Best,
Daniel
|