[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [datacenter] peak capacity planning for power vs. cooling
Okay. When looked at it that way...
100% of the power you consume is not used to generate heat. Its a high
number, but not 100%. So your peak heat requirements are always a little
lower than what you calculate.
As for your peak power requirements. The breakers you install themselves
have more than 20% headroom on them by design, but the cost difference
between 100 and 120A is negligible to install, so just install at least as
many as you have calculated. You will have some increases in resistance due
to power strips and power cords and things, but you have estimated pretty
conservatively in my opinion.
As for cooling, your biggest concern in a small room like that is not just
cooling (heat rejection) but airflow. You will almost certainly have a
problem with air circulation depending on how you are expecting heated air
to move. This would be the biggest concern I have if I were you.
DJ
> -----Original Message-----
> From: Ofer Inbar [mailto:[email protected]]
> Sent: Wednesday, January 15, 2003 5:28 PM
> To: datacenter
> Subject: Re: [datacenter] peak capacity planning for power vs. cooling
>
>
> I'm talking about a small machine room, of the "here, take this room
> and use it to put servers in, it's the only room we have available"
> variety. We figure out what machines are going to go in there, add
> the required number of circuits and the cooling needed, and install.
>
> I don't understand how "watts per square foot" applies in this case,
> to give any different answers about cooling than simply calculating
> the total and working from that.
>
> However, my real question was about the difference between capacity
> planning for power and cooling. I've seen one comment saying that I
> don't really need to have enough power for simultaneous peak. Anyone
> / anything else? How about cooling? Did the estimated calculations I
> presented seem to make sense?
>
> --
> -- Cos (Ofer Inbar) -- [email protected] 781-273-2380
> -- Permabit, Inc. -- [email protected] 617-252-9600
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]