ARM Challenging Intel in the Server Market: An Overview
by Johan De Gelas on December 16, 2014 10:00 AM ESTPerformance
The first SPECIntRate 2006 estimates were published by "CPU meister" Andreas Stiller. If we combine his findings with what we know and what is available at SPEC.org, we get the benchmark graph below.
Intel's own published SPECintrate scores are up to 20% higher, so at first sight the ARM competition is not there yet. However, we prefer to show the "lower numbers" as they have not been benchmarked with masterfully set ICC configuration settings.
The most aggressive architecture, the X-Gene, is quite a bit slower than the Xeon E3-1230L. The latter needs about 40W per node (SoC + chipset), while an X-Gene node would need almost 60W. AppliedMicro really needs the 28nm 2.8GHz X-Gene 2, which apparently can offer a 50% better performance per watt increase in SPECintrate 2006.
However, we have shown you that while SPECintrate 2006 is the standard often used, popular with most CPU designers, analysts and academic researchers, it is a pretty bad predictor of server performance. We should not discount the chances of the server ARM SoCs too quickly. A mediocre SPECint SoC can still perform well in server applications.
Facebook
Twitter
RSS
78 Comments
View All Comments
patrickjchase - Thursday, December 18, 2014 - link
It's been a while since I worked on this stuff, but I don't think that the statement that "CCN is very comparable to the ring bus found inside all Xeon processors beginning with Sandy Bridge" is quite right.CCN
patrickjchase - Thursday, December 18, 2014 - link
Finishing my comment:CCN
stefstef - Wednesday, December 17, 2014 - link
the idea of having an energy efficient design certainly will pay off. nvidia and samsung showed that having i.e. 4 cores and a fifth core dedicated to the energy management can be a good low cost solution. i dont often read the articles at anandtech because they are usually boring. although i am happy to place a coment here. arm rules in certain fields but in a couple of years only because intel will allow them to do so. every company needs a room to live in. another american breakfast for the chinese who will get their share in the processor market as well.milli - Thursday, December 18, 2014 - link
I don't understand how ARM is suddenly going to succeed while MIPS and PowerPC have already tried and failed. I feel that ARM is more of a market trend than anything else (in the server market).Even the current ARM server SOC manufacturers have already tried to penetrate the server market. Cavium and Broadcom already had custom designed low-power MIPS SOCs. IBM, Applied Micro and Freescale have had a bunch of low-power PowerPC options.
By the time any of these products is released, Intel is going to have a better alternative thanks to their process advantage. No IT manager is going to manage to convince any of the corporate fat-cats that a huge overhaul is needed. Same story over again.
yuhong - Friday, December 19, 2014 - link
"Unfortunately their 16GB DIMMs will only work with the Atom C2000, leading to the weird situation that the Atom C2000 supports more memory than the more powerful Xeon E3."I think the reason is software related. More precisely, the Memory Reference Code (MRC).
intiims - Tuesday, December 30, 2014 - link
If You want to know something about External Hard Drives visit http://www.hddmag.com/adrian1987 - Monday, January 5, 2015 - link
Hi. The Haswell core can actually have a max IPC of 6 instructions per cycle using macro-fusion not 5 as listed here (assuming the code is ideal). It has 2 execution units that can handle fused ALU and branch instructions. Source: http://www.anandtech.com/show/6355/intels-haswell-...aaronjoue - Tuesday, April 7, 2015 - link
Here is the real micro server. http://www.ambedded.com.tw/pt_list.php?CM_ID=20140...http://wiki.ambedded.com.tw/index.php?title=MicroS...
7 & 21 nodes in a chassis
It support Ubuntu and open source Ceph.