Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: [10GBASE-T] Power Down mode



Title: Power Down mode

Pat,

It is clear that the 10Gb chip must be designed to withstand long periods of data transmission. I don’t know how much   “long” is. You say that it may be hours for data centers. Now let imagine a multiport (24 ports) switch. Two Phys back up disks, but others are idle. The switch can dissipate MaxPower * 24 or much less power if other Phys enter the Power Down mode. So at box level the benefit is large.

 

The transition time delay and synchronization are not trivial, but it is purely technical issue. We can discuss it widely after PAR is approved.

 

The expression “it seems it would be a long time before people were deploying 10 Gig in battery powered devices”  I see as optimistic, opposite to possible “people will never deploy 10 Gig in battery powered devices”. Indeed it makes take a long time and according to common opinion 10Gb Phys then will be power efficient. But even today when 1000BASE-T phys consume less than 1W, everybody is looking to reduce its power consumption.

So here we have one more tool for power saving, tool that may be efficient and useful for application we even have not thought about.

 

Regards,

 

Boris Fakterman

 

-----Original Message-----
From: pat_thaler@agilent.com [mailto:pat_thaler@agilent.com]
Sent:
Thursday, August 07, 2003 11:47 PM
To: Fakterman, Boris; stds-802-3-10gbt@ieee.org
Subject: RE: [10GBASE-T] Power Down mode

 

Boris,

 

I don't think one can extrapolate from laptop/desktop traffic patterns to data centers.

 

The concern for power levels includes: ability to dissipate the heat generated, ability to supply the power and reliability effect of higher temperatures due to power. The first factor depends on the power level averaged over minutes/hours rather than days or weeks. The second factor depends on average power over sub-second times - whatever the decoupling capacitors can smooth out.

 

There will be times when utilization is high for an extended period. Such times can be hours (think about big disk back-ups for instance).

 

A power down mode is mainly helpful in response to total energy use concerns (saving the batteries or the planet) for systems that are sporatically active. It might be useful for reliability concerns (reducing total time spent at high temperatures). It doesn't answer concerns about power handling capability.

 

Also, from data center customers we often get concerns about latency including latency when an active period begins. I expect that transitioning from a low power mode to full operation would take non-zero time. There are already concerns about other delays (e.g. TCP congestion window) when moving from inactive to active. I expect that there would be some delay for adaptation when starting up.

 

Given all the above, I don't think a power down mode is useful in answering technical feasibility issues of power consumption. There may be other reasons to consider it in the project, though it seems it would be a long time before people were deploying 10 Gig in battery powered devices.

 

Pat

 

 -----Original Message-----
From: Fakterman, Boris [mailto:boris.fakterman@intel.com]
Sent:
Wednesday, August 06, 2003 7:54 AM
To: stds-802-3-10gbt@ieee.org
Subject: [10GBASE-T] Power Down mode

Hello all,

Following the discussion regarding power, it looks like there is a consensus that the 10Gb Phy dissipated power will be very high at first silicon and relatively high at advanced future versions.

The average power is important for most problematic topics, such as thermal conditions, power source availability and so on.

The average power can be reduced by using the Power-Down mode. The transceiver does not transmit or receive data during significant periods of time. Instead of transmitting idle symbols while consuming full power, the system can enter the Power-Down mode. The transmitter power can be reduced by stopping the transmission, the receiver power can be reduced as only minimal receive functions will be active. The overall dissipated power during the Power-Done can be reduced significantly.

Of course there are algorithmic issues to solve, such as how to maintain the synchronization during the Power-Down mode, but these are technical problems that can be discussed and solved.

The average power with implemented Power Down mode depends on the length of idle periods.

The desktop/laptop PC transmits idles most of the time ( > 90% ?). I don't know what happens in data centers.

.If we can reduce even half of the dissipated power by the Power Down mode, it is worth to be considered.

Regards,

Boris Fakterman - Intel Communications Group, Israel

Tel: 972-4-865-6470, Fax: 972-4-865-5999

mailto:boris.fakterman@intel.com