Re:Achieving 11M IOPS and 66 GB/S IO on a Single ThreadRippe...
by girvo on Saturday January 30, 2021 @03:10AM
(#25967277)
Would swapping them out for Noctuas be difficult?
[ Parent
| on HN ]
Re:Achieving 11M IOPS and 66 GB/S IO on a Single ThreadRippe...
by cfn on Saturday January 30, 2021 @04:27AM
(#25967637)
Mostly yes because server chassis are very compact and sometimes use proprietary connectors and fans. Still, many people have done that with good results, have a look in YouTube to know which server models are best suited for that kind of customization.
Re:Achieving 11M IOPS and 66 GB/S IO on a Single ThreadRippe...
by alvern on Saturday January 30, 2021 @04:00AM
(#25967476)
I've not been successful trying this with HPE servers. Most server fans (Foxconn/Delta) run 2.8 amp or higher.
Not aware of any "silent" gaming grade fans that use more than 0.38 amps.
That's not even considering the CFM.
Re:Achieving 11M IOPS and 66 GB/S IO on a Single ThreadRippe...
by weehoo on Saturday January 30, 2021 @09:54AM
(#25969534)
Why would current be relevant here? Shouldn’t operating voltage be the only thing that matters?
Re:Achieving 11M IOPS and 66 GB/S IO on a Single ThreadRippe...
by sokoloff on Saturday January 30, 2021 @10:23AM
(#25969774)
Amps * Volts is power. Power is a proxy (a moderately good one) for air movement (a mix of volume/mass at a specific [back-]pressure).
It’s not likely that a silent 2W fan will move a similar amount of air as the stock 14W fans. The enterprise gear from HPE is pretty well engineered; I’m skeptical that they over-designed the fans by a 7x factor.
Operating voltage tells you “this fan won’t burn up when you plug it in”. It doesn’t tell you “will keep the components cool”.