result of this public bickering
is that it is difficult for chip customers to figure out who to
To settle this dispute, we will analyze each of these claims and provide
data to validate or debunk each of these claims.
slowest Intel Nehalem is faster
than the fastest AMD Shanghai
No, yes, and not anymore.
If we include the one dual-core Nehalem
variant, the answer is no because a lower clock speed dual-core Intel Nehalem
isn't going to beat the highest
clocked quad-core AMD Shanghai.
If Intel was only talking about all of the other quad-core Nehalem
models then yes, even the slowest
2 GHz Nehalem XEON model E5504 will
beat the fastest 2.7 GHz AMD Shanghai
Opteron model 2384. But this has changed recently as AMD has
released a 3.1 GHz model 2393SE Opteron which would likely be faster
slowest Nehalem quad-core processor
but still lag far behind most Nehalem
1a and 1b below shows the
latest and best performance scores published at SPEC.org as of
SPECint is a general purpose benchmark covering a lot of mainstream
applications while SPECfp represents a cross section of scientific and
engineering applications. SPECweb reflects web server
For comparison sakes, we'll also include the fastest Nehalem
Xeon 2.93 GHz X5570 which on average is nearly twice as
fast as the Opteron 2384. Since official public benchmarks
available at the time of this writing, table 1 below will not include
2393SE performance numbers. However, the performance gain of
2393SE will likely be no more than 15% faster than an Opteron 2384
which may be
enough to put it past the Intel E5504.
1a: SPEC CPU 2006 base performance results as of 4/27/2009
SPECfp data is especially
noteworthy because this was traditionally an AMD stronghold before the
of Intel's Nehalem processor with
QuickPath memory architecture. SPECfp along with everything
else is now
clearly dominated by Intel Nehalem.
1b: SPEC WEB 2005 performance results as of 4/27/2009
was another strong suit for
AMD before the arrival of Intel Nehalem.
Now a dual-socket Nehalem based
server can even outpace a four-socket Shanghai
based Opteron server.
launched the world's first
quad-core processor within a 40-watt thermal envelope and has the most
did not release a quad-core
processor within the 40-watt thermal envelope and they do not have the
energy efficient chip. While AMD never officially claimed
newest Opteron 2377HE chip runs inside a 40-watt thermal envelope, this
impression they have successfully marketed to the public. AMD
to disclose the actual Thermal Design Power (TDP) of their latest CPUs
media and has tried to state that AMD's Average CPU Power (ACP) metric
comparable to Intel's TDP metric. As a result, every news
product label, and every server vendor only show ACP numbers and not
thermal ceiling of the processors while Intel continues to use the TDP
which more accurately reflects peak power consumption.
also claims that Intel's
processors have a higher thermal ceiling which makes them less suitable
dense data center deployments, but even this claim is false for most Nehalem
models. The most
comprehensive and standardized server power efficiency metric to date
and it shows a wide range of power consumption metrics across a variety
levels. Table 2 below will show a comparison of the most
efficient AMD Shanghai based
servers against the most
efficient Intel Nehalem based
servers. SPECpower_ssj2008 is so interesting because it not
only tells us
the actual measured idle and peak power consumption under a server side
load, but it also tells us the server's server side java
"score" is determined by an average measurement of performance per
watt across all the workloads. SPECpower scores for the 2.0
were not available, so the closest model number with the most similar
speed and power consumption characteristics were used in table 2.
2: SPECpower_ssj2008 results as of 4/27/2009
we can see from the official
SPECpower_ssj2008 results above, AMD's "High Efficiency" (HE)
products actually consume substantially more power than Intel's low
products despite the fact that AMD advertises the lowest power
consumption. The idle power consumption gap is even larger
significant since most servers spend most of their time low utilization
states. It's clear that the million plus transistors Intel
power management on the Nehalem
is paying off.
these measured peak power
numbers don't reflect the maximum power that each system can draw under
workload, it does represent the peak power consumption under this
type of workload which is generally not as intensive as a High
Computing (HPC) workload used in scientific and engineering
What these numbers do confirm is that AMD's claim that their ACP metric
similar to Intel's TDP metric is clearly false and that AMD's TDP
these processors are far more comparable to Intel's TDP.
also noteworthy that this
benchmark also shows lower clocked Intel processors performing more
java operations per second (SSJ_OPS) than the higher clocked AMD
which further reinforces the data from Table 1.
equivalent Intel Nehalem
server costs almost twice as much as an equivalent AMD Opteron Server.
This claim is false because there are no equivalent (in
performance or power consumption) AMD Opterons compared to Intel Nehalem
servers. While AMD
servers are the cheapest, a low end Nehalem
server can deliver more performance at a lower price. Table 3
server prices from HP with the base configuration plus second processor.
3: Server price comparison
appears that AMD is already
heavily adjusting their processor prices to the major server makers
the unrealistic list prices that were originally set before the arrival
Intel Nehalem. In fact if
only went by the official processor list prices, we would expect the
servers to cost much more than they do in table 3. It is
normal to see a
lag in price adjustment immediately after a new chip lunch when
haven't all caught on to the new products yet and when inventory is
working its way into the channel. However, we should expect
price cuts to come soon because there's no reason that a server based
Opteron 2384 should fetch more money than a server based on Intel Nehalem
server given the benchmark
on the high end multi
processor server market
implications of Intel Nehalem on
the server market for the
near term are profound. It not only threatens AMD in the
market, but it threatens AMD's four-socket market and even begins to
on Intel's four-socket Dunnington
servers. When you can buy a two-socket Intel Nehalem
server at a fraction of the price and exceed the
performance of four-socket AMD servers and come close to Intel Dunnington,
it takes the thunder out of
four-socket servers. Table 4a and 4b shows two-socket Intel Nehalem
servers beating Opteron
four-socket servers in SAP and Virtualization performance.
4a: SAP performance *
4b: VMware virtualization performance *
@ 14 Tiles
@ 16 Tiles
results are stunning because
at no time in the last 5 years has the performance gap between Intel
been so large. This situation won't change until AMD launches
6-core Istanbul Opteron processors
expected by June 2009 and Intel launches its 8-core Nehalem-EX
multi-socket severs expected by the end of Q4 2009. AMD
6-core Istanbul will narrow their
deficit not enough to close it against Intel current Nehalem-EP
4-core processors. However, Intel's Nehalem-EX
8-core processor is expected
to widen Intel's performance lead by a significant margin by the end of
year. But because Nehalem-EX
won't be seen in most mainstream two-socket servers, Istanbul
will allow AMD to raise their average selling price to a
appears that the server market
between Intel and AMD is now mirroring the desktop chip wars.
customers have a choice of buying low-cost value servers based on AMD Shanghai
processors or they can buy
premium priced Intel-based servers for better performance and energy
efficiency. However, the value hunter should wait a few more
weeks when Shanghai servers settle
down to post-Nehalem prices.
* All results and prices as of
quote: Chips too slow for what?
quote: The Atom is the current hot item right now, and that is anything but powerful.
quote: Only some server functions need exceptionally fast CPUs.
quote: I grabbed a 4850e which featured low power
quote: Intel holding back newer products so as not to canibalize their existing sales.
quote: Lag in performance? those PII's seem to be doing a bang up job againts intels 7,8, and 9x lines which is where we see the bulk of Intel's desktop sales.