Design Con 2015
Breaking News
Comments
Newest First | Oldest First | Threaded View
Page 1 / 2   >   >>
TarraTarra!
User Rank
CEO
Re: Intel, ARM Do Server Wrestling at ISSCC
TarraTarra!   11/7/2013 4:19:44 PM
NO RATINGS
NOC vs. Ring:

 

Disagree that interconnect power is at a scale that compares with CPU power. CPUs are very much the highest power consuming circuits on the die by a huge margin. The interconnect discussions of NOC vs. Ring again is confusing things. NOCs are used to connect to peripheral interfaces and IP. NOCs are not coherent by definition. The Rings that Intel uses are for the coherent CPU interconnect.

Reducing the CPU power has the biggest bang for the buck while still maintaining performance. Adding CPUs to the die increases the power linearly but the interconnect power increase is less than linear.

 

 

sw guy
User Rank
CEO
Re: Intel, ARM Do Server Wrestling at ISSCC
sw guy   11/7/2013 7:33:54 AM
NO RATINGS
That optimization is both possible and non trivial explains why at least 2 NoC IP companies are able to live.


BTW, some people would argue that a ring topology is kinda network. In that case, we should say it is one with (good) static optimization of routing.


As usual, there is no "one seizes all" solution. There are compromises to be done, and final choices depend on various parameters.

 

rick merritt
User Rank
Author
Re: Intel, ARM Do Server Wrestling at ISSCC
rick merritt   11/6/2013 10:55:05 AM
NO RATINGS
A microserver tech lead at Dell told me they did comparisons that showed Intel's Silvermont (64-bit 22nm Atom) core is about on par in perf and perf/W with the ARM A15 (32-bit 28 nm core).

rick merritt
User Rank
Author
Re: Intel, ARM Do Server Wrestling at ISSCC
rick merritt   11/6/2013 10:53:21 AM
NO RATINGS
@GSMD: Thanks for the insights on chip interconnects! I would not have thought they were such power hogs either.

Wilco1
User Rank
CEO
Re: Ivytown & ARM don't compete
Wilco1   11/6/2013 8:58:28 AM
NO RATINGS
"the point is that pie is already divided and the ARM is going after a piece that is already covered by Intel, which is already on its 2nd-generation Atom microserver chip before ARM is even out of the gate."

That's an interesting rewrite of history... Calxeda has had its ARM servers out for well over a year now, and that was before Intel even announced Centerton, let alone shipped it! Note Calxeda has its 2nd generation out as well.

I also don't agree that the x86 penalty is low - if that were true then why is AMD having such a hard time keeping up with Intel while a dozen of small outfits can design fast and efficient ARM cores which are challenging Intel? Even Intel took a very long time to come up with an Atom replacement, and it ended up being a simple 2-way core (as 3/4-way is too power hungry on x86).

Wilco1
User Rank
CEO
Re: Intel, ARM Do Server Wrestling at ISSCC
Wilco1   11/6/2013 8:24:25 AM
NO RATINGS
What A7 are you talking about? And what does transistor physiscs or ALU layout have got to do with this?

If you didn't understood it, in simple terms: a limited 2-way out-of-order CPU is never going to beat a 4-way fully OoO one. Better transistor physics does not help there.

HangLai
User Rank
Rookie
Re: Intel, ARM Do Server Wrestling at ISSCC
HangLai   11/6/2013 8:10:20 AM
NO RATINGS
Wilco1:  Please show your evidence that shows A7 performed better than Intel's, either through the transistor physics or the architecture ALU layout in the chip.  Please don't put out any alleged information that is non-scientific and no basis.

Wilco1
User Rank
CEO
Re: Intel, ARM Do Server Wrestling at ISSCC
Wilco1   11/6/2013 7:00:08 AM
NO RATINGS
Remember this is 18W for a 3GHz octo core at 40nm. A 22nm C2750 runs at 2.4GHz, uses 20W and cannot achieve anywhere near the same performance (X-Gene should have better than Cortex-A57 performance, while we already know Silvermont is slower than Cortex-A15 clock for clock).

So it looks like Avoton will be beaten by a huge margin on performance and power efficiency despite having the advantage of 2 process generations. Now imagine a next-generation X-gene at TSMC 20nm...

 

sw guy
User Rank
CEO
Re: Intel, ARM Do Server Wrestling at ISSCC
sw guy   11/6/2013 6:52:14 AM
NO RATINGS
First time I hear of NoC as source of high power consumption.

Maybe it depends on actual topology and/or implementation ?

dynamited
User Rank
Rookie
Re: Intel, ARM Do Server Wrestling at ISSCC
dynamited   11/5/2013 5:52:53 PM
NO RATINGS
Did I read correctly the ARM offering was going to be 18watts (4x4.5)? Not exactly low power. Any anaylsis on building more chassis to house more processors that are less capable from a reliability point of view?

Page 1 / 2   >   >>


Flash Poll
Top Comments of the Week
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Life
Frankenstein's Fix, Teardowns, Sideshows, Design Contests, Reader Content & More
Max Maxfield

Book Review: Deadly Odds by Allen Wyler
Max Maxfield
11 comments
Generally speaking, when it comes to settling down with a good book, I tend to gravitate towards science fiction and science fantasy. Having said this, I do spend a lot of time reading ...

Martin Rowe

No 2014 Punkin Chunkin, What Will You Do?
Martin Rowe
1 Comment
American Thanksgiving is next week, and while some people watch (American) football all day, the real competition on TV has become Punkin Chunkin. But there will be no Punkin Chunkin on TV ...

Rich Quinnell

Making the Grade in Industrial Design
Rich Quinnell
13 comments
As every developer knows, there are the paper specifications for a product design, and then there are the real requirements. The paper specs are dry, bland, and rigidly numeric, making ...

Martin Rowe

Book Review: Controlling Radiated Emissions by Design
Martin Rowe
1 Comment
Controlling Radiated Emissions by Design, Third Edition, by Michel Mardiguian. Contributions by Donald L. Sweeney and Roger Swanberg. List price: $89.99 (e-book), $119 (hardcover).