nanog mailing list archives

Re: "Hypothetical" Datacenter Overheating


From: Bryan Holloway <bryan () shout net>
Date: Tue, 16 Jan 2024 10:14:10 +0100


On 1/15/24 23:11, Martin Hannigan wrote:


On Mon, Jan 15, 2024 at 4:10 PM Jay Hennigan <jay () west net <mailto:jay () west net>> wrote:

    On 1/15/24 10:37, Pennington, Scott wrote:
     > yes but.... it has been -8 in Chicago plenty of times before this.
     >   Very interested in root cause...

    Absolutely. My point was that claiming "Global warming" isn't going to
    fly as an excuse.


+1

Is their design N+1?

https://www.equinix.com/data-centers/americas-colocation/united-states-colocation/chicago-data-centers/ch1 
<https://www.equinix.com/data-centers/americas-colocation/united-states-colocation/chicago-data-centers/ch1>

We're not smashing temp records in Chicago. At least it doesn't seem so when you look across historical data:

https://www.weather.gov/lot/Chicago_Temperature_Records <https://www.weather.gov/lot/Chicago_Temperature_Records>

HTH,

-M<

I was at that "hypothetical" location once when it was -17º F ...

No issues then ...

I really have to wonder how six chillers all failed at once.


Current thread: