Serial Port Memory Technology

Article: 8-Socket Commodity Servers: Flourish or Perish?
By: David Hess (davidwhess.delete@this.gmail.com), March 21, 2010 3:32 am
Room: Moderated Discussions
Michael S (already5chosen@yahoo.com) on 3/15/10 wrote:
---------------------------
>slacker (s@lack.er) on 3/14/10 wrote:
>---------------------------
>>longtimelurker (rwt@nospam.maibaums.net) on 3/14/10 wrote:
>>---------------------------
>>>Which brings me to the actual point I'm trying to make: Isn't the number of sockets
>>>an antiquated metric? What you are really interested in with those big boxes is
>>>the total number and throughput of the memory channels and that they're reasonably low latency from any core.
>>
>>I agree, in a limited sense, that number of sockets is an antiquated metric. This
>>is because, in The Old Way™, multiple sockets was the only way to have multiple
>>cores in general purpose machines. Thus, performance scaling across many sockets
>>was an important metric. In The New Way™, you get multiple cores in a single socket,
>>so why care about multi-socket scaling?
>>
>>All of the problems which had useful parallelism across 2 - 64 cores will eventually
>>be shoe-horned on to a single chip. For these problems, socket count will eventually
>>be totally irrelevant (because they'll run on single-socket, many-core processors).
>>
>>However, you can only get so many pins on a regular-sized package. Since you need
>>to scale memory bandwidth with the number of cores, it will become increasingly
>>difficult to maintain some sort of balance between bandwidth:FLOPs. It's not very
>>useful to integrate a large number of cores on a single chip if you can't get enough
>>data in and out of the chip. Thus, multi-socket systems will be the primary method
>>of increasing aggregate memory bandwidth in many-core systems.
>>
>
>Thinking about it, aggregate memory bandwidth delivered as lot of independent streams
>is something that money+power+latency could buy, even within physical limitations
>of a single socket. Imagine a true serial solution, not glorified parallel, like
>HT3/QPI/FBD, not multilane serial, like IB/PCIe/XAUI, but a true serial like 10GBASE-T
>just, say 25% slower and reaching 100cm instead of 100m. With a lot of engineering
>effort it should be possible to build such phy under 0.5W. 256 channels like that
>should comfortably fit in 1500-pin package. With 256B packets each channel would deliver 900 MB/s with 90% efficiency.
>At 64 cores there are 4 channels per core = 3.2 GB/s as measured by Stream. That's
>close to what we have today in Nehalem-EP, better than Shanghai and significantly better than Nehalem-EX and Istanbul.
>
>So, bandwidth, by itself, is solable. High bandwidth of individual channel and/or
>low latency are much harder nuts to crack. But, then again, for non-NUMA-friendly
>access pattern in 16-socket system those characteristics are not particularly great
>either, although 2-4 times better than the crazy single socket from my above thought experiment.

They say this is intended for mobile and portable devices but it seems a lot like what you are describing. I have not registered for any of the technical papers yet.

Serial Port Memory Technology:

http://www.spmt.org
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
New article: 8 socket commodity serversDavid Kanter2010/03/09 10:27 AM
  New article: 8 socket commodity serversVincent Diepeveen2010/03/09 01:46 PM
    number of sockets is wrong metric (was: New article: 8 socket commodity servers)longtimelurker2010/03/14 05:13 AM
      number of sockets is wrong metric (was: New article: 8 socket commodity servers)EduardoS2010/03/14 05:34 AM
      number of sockets is wrong metric (was: New article: 8 socket commodity servers)Wes Felter2010/03/14 10:33 AM
        Magny-CoursMax2010/03/14 04:56 PM
          Magny-Coursanonymous2010/03/14 06:33 PM
            Magny-Courslongtimelurker2010/03/15 02:54 AM
      number of sockets is wrong metric (was: New article: 8 socket commodity servers)Vincent Diepeveen2010/03/14 11:31 AM
        number of sockets is wrong metric (was: New article: 8 socket commodity servers)longtimelurker2010/03/14 01:37 PM
          number of sockets is wrong metric (was: New article: 8 socket commodity servers)Vincent Diepeveen2010/03/15 11:36 AM
      number of sockets is wrong metric (was: New article: 8 socket commodity servers)David Kanter2010/03/14 11:56 AM
        Bad mathDavid Kanter2010/04/01 01:24 AM
      number of sockets is wrong metric (was: New article: 8 socket commodity servers)slacker2010/03/14 02:51 PM
        number of sockets is wrong metric (was: New article: 8 socket commodity servers)Michael S2010/03/15 05:05 AM
          number of sockets is wrong metric (was: New article: 8 socket commodity servers)slacker2010/03/15 01:02 PM
            Memory interfacesDavid Kanter2010/03/15 01:17 PM
              Memory interfacesslacker2010/03/15 09:08 PM
                Patents on tiny components vs. large, complex thingsmpx2010/03/15 11:41 PM
                  Patents on tiny components vs. large, complex thingsRichard Cownie2010/03/16 05:58 AM
                    Patents on tiny components vs. large, complex thingsMS2010/03/17 05:42 PM
                      Patents on tiny components vs. large, complex thingsa reader2010/03/18 08:45 PM
          Serial Port Memory TechnologyDavid Hess2010/03/21 03:32 AM
  New article: 8 socket commodity serversMichael S2010/03/09 03:13 PM
    New article: 8 socket commodity serverstheluketaylor2010/03/09 05:32 PM
    New article: 8 socket commodity serversJesper Frimann2010/03/09 10:35 PM
    New article: 8 socket commodity serversDavid Kanter2010/03/10 12:38 AM
      New article: 8 socket commodity serversTim2010/03/16 08:44 AM
  New article: 8 socket commodity serversanon2010/03/09 06:59 PM
    New article: 8 socket commodity serversDavid Kanter2010/03/09 11:06 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?