New article - The Common System Interface: Intel's Future Interconnect

Article: The Common System Interface: Intel's Future Interconnect
By: Jonathan Kang (johnbk.delete@this.gmail.com), September 25, 2007 8:05 am
Room: Moderated Discussions
8B/10B Latency (dwhess@banishedsouls.org) on 9/22/07 wrote:
---------------------------
>David Kanter (dkanter@realworldtech.com) on 9/22/07 wrote:
>---------------------------
>>Jigal (jigal2@gmail.com) on 9/22/07 wrote:
>>---------------------------
>>>Hi there,
>>>
>>>Being a newbie I throw myself at the mercy of the forum.
>>
>>Welcome - I think you'll find we're a pretty merciful bunch.
>
>Burn him! jk :)
>
>>>Small question - how come they didn't leverage the PCI >Express and needed a new
>>>bus (excuse me, p2p interconnect) altogether?
>>
>>PCI Express isn't coherent, it's also fairly high latency since it uses 8B/10B clock encoding.
>
>How much latency does 8B/10B encoding contribute? If I am reading this correctly,
>Lattice has a programmable logic implementation optimized for throughput with only
>2 clocks of latency on the encoder and 3 clocks on the decoder when working exclusively with serial bit streams:
>
>http://www.latticesemi.com/dynamic/view_document.cfm?document_id=5653
>
>CSI would require a factor of 60 speed up of course. I admittedly have never had
>to deal with the logic design for high speed 8B/10B. Could the shift from a PLD
>design to a full custom one yield that large a difference?
>
>I suspect there is not a good reason to use 8B/10B encoding and sacrifice 20% of
>your throughput where a clock can be made available unless you want to support AC
>coupling like Hypertransport 3 where they do list lower throughput and higher latency
>as a disadvantage for this type of operation.
>

It's not the encode/decode latency, that can easily be done one clock (assuming no pipelining) with fast enough gates. It's the clock recovery. A PLL to recover a serial clock from the data-stream takes hundreds of cycle times. This effectively gives you two options:

1. Keep the link active at all times such that the receive PLL never loses sync. This burns power. This provides the low-latency requirement though.
2. Turn the link off, and when it turns on, spend hundreds of cycle-times to recover the clock from the data-stream.

Then there's issues of synchronization to the local clock of the receiver, which generally takes a few clock cycles but I suppose you'd have to do that with a scheme like CSI as well.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/08/28 02:16 AM
  New article - The Common System Interface: Intel's Future InterconnectVincent Diepeveen2007/08/28 06:02 AM
  New article - The Common System Interface: Intel's Future InterconnectRichard Cownie2007/08/28 11:28 AM
    New article - The Common System Interface: Intel's Future InterconnectVincent Diepeveen2007/08/31 12:44 PM
      New article - The Common System Interface: Intel's Future InterconnectRichard Cownie2007/08/31 09:53 PM
        New article - The Common System Interface: Intel's Future InterconnectVincent Diepeveen2007/09/01 03:21 AM
          Adding layers can simplify designPaul A. Clayton2007/09/01 08:39 AM
          New article - The Common System Interface: Intel's Future InterconnectMichael S2007/09/02 03:25 AM
        New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/14 01:47 PM
    New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/09/14 09:47 PM
  New article - The Common System Interface: Intel's Future InterconnectPaul2007/08/28 12:04 PM
    New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/08/28 01:43 PM
      New article - The Common System Interface: Intel's Future InterconnectJoe Chang2007/08/28 07:17 PM
        New article - The Common System Interface: Intel's Future InterconnectJoe Chang2007/08/29 05:27 PM
  Thanks for the workWouter Tinus2007/08/28 01:33 PM
    Thanks for the workmac2007/08/29 01:44 PM
  New article - The Common System Interface: Intel's Future InterconnectHerbert Hum2007/08/28 02:22 PM
    ThanksDavid Kanter2007/08/28 05:13 PM
  Many thanks, very, very interesting! (NT)Cameron Jack2007/08/29 02:51 AM
  very nice article + memory ctrl integrationMarcin Niewiadomski2007/08/29 12:46 PM
    very nice article + memory ctrl integrationDavid Kanter2007/09/14 09:50 PM
      very nice article + memory ctrl integrationMarcin Niewiadomski2007/09/16 09:48 PM
  Coherency: Forwarding and OwnedPeter Gerdes2007/08/29 03:11 PM
    Coherency: Forwarding and OwnedDavid Kanter2007/08/29 07:29 PM
      Coherency: Forwarding and Ownednick2007/08/29 08:03 PM
        Coherency: Forwarding and OwnedDavid Kanter2007/08/30 12:08 AM
      Coherency: Forwarding and OwnedMichael S2007/08/30 02:17 AM
        Coherency: Forwarding and OwnedDavid Kanter2007/08/30 08:31 AM
      Coherency: Forwarding and OwnedPeter Gerdes2007/08/30 12:46 PM
        Coherency: Forwarding and OwnedDavid Kanter2007/08/30 02:46 PM
          Coherency: Forwarding and OwnedPeter Gerdes2007/08/30 08:03 PM
            Coherency: Forwarding and OwnedDavid Kanter2007/09/14 09:44 PM
              Node Interleaveunknown2007/09/15 04:14 AM
                Node InterleaveDavid Kanter2007/09/15 08:50 AM
                  Node InterleaveHoward Chu2007/09/16 01:14 PM
              Coherency: Forwarding and OwnedPeter Gerdes2007/09/16 01:50 PM
                Coherency: Forwarding and OwnedDavid Kanter2007/09/16 05:34 PM
                  Coherency: Forwarding and OwnedEduardoS2007/09/16 05:52 PM
                  Coherency: Forwarding and OwnedJonathan Kang2007/09/17 06:16 AM
                  Coherency: Forwarding and OwnedMatthias2007/09/17 07:59 AM
                    Coherency: Forwarding and Owned - additionMatthias2007/09/17 08:01 AM
                Coherency: Forwarding and Ownedanonymous2007/09/17 10:15 AM
                  Coherency: Forwarding and OwnedPeter Gerdes2007/09/17 01:44 PM
  New article - The Common System Interface: Intel's Future InterconnectMr. Camel2007/08/30 04:16 PM
    New article - The Common System Interface: Intel's Future InterconnectMichael S2007/08/31 02:11 AM
      New article - The Common System Interface: Intel's Future InterconnectMr. Camel2007/08/31 04:13 AM
        New article - The Common System Interface: Intel's Future InterconnectMichael S2007/08/31 04:24 AM
        New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/08/31 06:39 AM
          New article - The Common System Interface: Intel's Future InterconnectMichael S2007/08/31 07:53 AM
      New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/08/31 06:41 AM
        New article - The Common System Interface: Intel's Future InterconnectMichael S2007/08/31 07:36 AM
          New article - The Common System Interface: Intel's Future InterconnectMr. Camel2007/08/31 09:36 AM
  Thanks and excellent work!Jack A.2007/08/30 08:41 PM
  Lamport's TLAKonrad Schwarz2007/09/02 02:57 AM
    Lamport's TLADavid Kanter2007/09/02 08:55 PM
    Lamport's TLABrannon2007/09/03 08:12 AM
      Lamport's TLAKonrad Schwarz2007/09/18 11:21 AM
        Lamport's TLABrannon2007/09/18 02:58 PM
  New article - The Common System Interface: Intel's Future InterconnectJosé Javier Zarate2007/09/09 05:01 PM
  New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/16 06:42 PM
    Remote prefetchDavid Kanter2007/09/17 08:51 AM
  New article - The Common System Interface: Intel's Future InterconnectJigal2007/09/22 03:39 PM
    New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/09/22 05:35 PM
      New article - The Common System Interface: Intel's Future Interconnect8B/10B Latency2007/09/22 07:16 PM
        New article - The Common System Interface: Intel's Future Interconnectanon2007/09/22 09:05 PM
          New article - The Common System Interface: Intel's Future InterconnectDavid W. Hess2007/09/22 09:50 PM
            Clocking in CSIDavid Kanter2007/09/23 09:46 AM
              Hypertransport 3 AC CouplingDavid W. Hess2007/09/23 10:32 AM
                Hypertransport 3 AC Couplinganon2007/09/23 10:53 AM
                  Clocking lanesDavid Kanter2007/09/23 11:51 AM
        New article - The Common System Interface: Intel's Future InterconnectDavid Kanter2007/09/22 09:34 PM
          New article - The Common System Interface: Intel's Future InterconnectDavid W. Hess2007/09/22 10:10 PM
            New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/25 08:15 AM
          New article - The Common System Interface: Intel's Future InterconnectMichael S2007/09/23 02:06 AM
            New article - The Common System Interface: Intel's Future InterconnectDavid W. Hess2007/09/23 04:41 AM
            Serialization delayDavid Kanter2007/09/23 09:57 AM
              Serialization delayMichael S2007/09/23 12:20 PM
                Serialization delayDavid Kanter2007/09/23 01:43 PM
                  Serialization delayMichael S2007/09/24 01:40 AM
                    Serialization delayMichael S2007/09/24 05:28 AM
                    Serialization delayAaron Spink2007/09/24 01:19 PM
                      Serialization delayMichael S2007/09/25 04:38 AM
                        Serialization delayJonathan Kang2007/09/25 09:10 AM
                          Serialization delayDavid W. Hess2007/09/26 01:22 AM
                        Serialization delayAaron Spink2007/09/25 01:13 PM
                          Thank you (NT)Michael S2007/09/25 01:53 PM
                Serialization delayJonathan Kang2007/09/25 08:26 AM
                  Serialization delayMichael S2007/09/25 02:57 PM
                    Serialization delayJonathan Kang2007/09/26 06:24 AM
                      Serialization delayDavid W. Hess2007/09/26 07:39 AM
                        Serialization delayJonathan Kang2007/09/26 10:56 AM
                          Serialization delayDavid W. Hess2007/09/27 03:21 AM
                            Serialization delayJonathan Kang2007/09/27 05:36 AM
                              Serialization delayDavid W. Hess2007/09/27 06:31 PM
                      Serialization delayrwessel2007/09/26 02:26 PM
                        Serialization delayJonathan Kang2007/09/27 08:16 AM
                          Serialization delayrwessel2007/09/27 01:20 PM
                            Serialization delayJonathan Kang2007/09/28 05:38 AM
                              Serialization delayrwessel2007/09/28 02:00 PM
                                Serialization delayJonathan Kang2007/10/01 08:07 AM
                                  Cache coherent latencyDavid Kanter2007/10/01 08:20 AM
                                    Cache coherent latencyblaine2007/10/01 11:36 AM
                                      Critical word first on coherent interconnectsDavid Kanter2007/10/01 12:10 PM
                                        Does ccHT do critical word first?blaine2007/10/02 08:10 AM
                                    Cache coherent latencyJonathan Kang2007/10/01 01:34 PM
                                      Cache coherent latencyDavid Kanter2007/10/01 02:13 PM
                      Serialization delayMichael S2007/09/28 05:32 AM
                        Serialization delayanonymous2007/09/28 11:25 AM
                          Serialization delayMichael S2007/09/29 10:06 AM
        New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/25 08:05 AM
      New article - The Common System Interface: Intel's Future Interconnectjigal2007/09/23 01:37 PM
        CSI, PCI and HTDavid Kanter2007/09/23 01:46 PM
        New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/25 08:39 AM
          New article - The Common System Interface: Intel's Future Interconnectjigal2007/09/25 03:16 PM
            New article - The Common System Interface: Intel's Future InterconnectMichael S2007/09/26 04:14 AM
              New article - The Common System Interface: Intel's Future InterconnectAnonymous2007/09/26 10:41 AM
                New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/26 10:59 AM
            New article - The Common System Interface: Intel's Future InterconnectJonathan Kang2007/09/26 06:48 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell avocado?