By: Michael S (already5chosen.delete@this.yahoo.com), September 28, 2007 5:32 am
Room: Moderated Discussions
Jonathan Kang (johnbk@gmail.com) on 9/26/07 wrote:
---------------------------
>
>From what I've read of the various Rambus patents that seem to relate to this (7271623
>and 7269706 are interesting in particular), it seems that they've been able to
>cut power consumption by improving upon just about every component in a transmission
>link in an iterative fashion. For instance, they've made a very sensitive receiver
>that does selective amplification based on a reference clock to cut on transmission pre-emphasis power, etc.
>
I got the same impression. Not a one big invention but attention to details on all levels and designing in complex.
For example, data available from CDR block helps to build simpler and more precise equalizer. Then better equalizer on the Rx side enables Tx side without pre-emphasis.
The fox knows many things, but the hedgehog knows one big thing. Rambus is a fox.
>This does not, however, get around the fundamental issue of the link having to
>be active at all times. Remember that whatever power figures given for parallel
>links such as CSI are only the active power figures. When comparing it to a serialized
>link, even a low-powered one like the one Rambus has, one has to take into account average power.
I think that at least for current server environment they reached "good-enough" point.
For dual socket system you want approximately 500 Gbit/s of interconnect bandwidth.
500 Gbit/s * 2 mW*s/Gbit = 1W.
Taking into account that idle power consumption of today's most economical dual-socket server exceeds 150W it seems that we can live with 1W constantly consumed by interconnects.
Quad-socket server would need 5-6 times more of interconnect bandwidth but even 6 Watt doesn't sound frightening for such relatively big box.
---------------------------
>
>From what I've read of the various Rambus patents that seem to relate to this (7271623
>and 7269706 are interesting in particular), it seems that they've been able to
>cut power consumption by improving upon just about every component in a transmission
>link in an iterative fashion. For instance, they've made a very sensitive receiver
>that does selective amplification based on a reference clock to cut on transmission pre-emphasis power, etc.
>
I got the same impression. Not a one big invention but attention to details on all levels and designing in complex.
For example, data available from CDR block helps to build simpler and more precise equalizer. Then better equalizer on the Rx side enables Tx side without pre-emphasis.
The fox knows many things, but the hedgehog knows one big thing. Rambus is a fox.
>This does not, however, get around the fundamental issue of the link having to
>be active at all times. Remember that whatever power figures given for parallel
>links such as CSI are only the active power figures. When comparing it to a serialized
>link, even a low-powered one like the one Rambus has, one has to take into account average power.
I think that at least for current server environment they reached "good-enough" point.
For dual socket system you want approximately 500 Gbit/s of interconnect bandwidth.
500 Gbit/s * 2 mW*s/Gbit = 1W.
Taking into account that idle power consumption of today's most economical dual-socket server exceeds 150W it seems that we can live with 1W constantly consumed by interconnects.
Quad-socket server would need 5-6 times more of interconnect bandwidth but even 6 Watt doesn't sound frightening for such relatively big box.
