If you like pfsense as software but not the community have a look on the OPNSense fork.
I use it since a year everywhere where I used pfsense before and am very happy with it.
Gesendet von meinem BLA-L29 mit Tapatalk
If you only have 1-2 minutes of runtime you have to shutdown your stack every time power drops. Also you would have to wait 3-4 hours until the batteries are recharged before powering on again, because there could be a 2nd drop. Better go for about 5x the time you need to shutdown.
Achieves 17GB/s read and 10GB/s write using 8x8 port SAS2. (With 80FMODs, this beast has 96!)
Sun Storage F5100 Flash Array with PCI-Express SAS-2 HBAs Achieves Over 17 GB/sec Read
Theres also a 2nd one on ebay with only 20 FMODs, which would be ideal as spare parts (599$).
SUN F5100 with 96x24GB SLC SSD.
The bigger cousin of the Oracle F40.
1.600.000 IOPS Read
1.200.000 IOPS Write
Only 1 available.
Have Fun and please share some benchmarks!
Sun Storage F5100 2TB SAS Flash Array 96 x 24GB 371-4095-05 TESTED w/WARRANTY! | eBay
You can use a HP P420 Controller.
The problem is, that the drive temperature is read wrong by the P410i and the fans spin up to 100%.
This shouldn't happen with a P420 and it will also be a reasonable upgrade (P420 can be set to passthrough/JBOD if you use ZFS or similar)
Google has your ass with coreboot-NERF and u-root. With their fully open-source replacement of the UEFI they brought down the boot time of a winterfell node (OCP WiWynn) from 7 minutes to 20s.
They already have 4 motherboards supported (unfortunately I couldn't find out which).
Most probably there is some raid config on the end of the drives which BSD recognizes using its geom_raid module.
The easiest way to get rid of it is clearing it the same way it was created.
Alternatively nuke the whole drive.
Just add a few VDI Desktops and workers on that host and CPU will be the limiting factor.
We have a client with about 1700 persistent VDI Desktops on 160 Cores. It all went well for a few months (<25% CPU load on a work day)...
It seems CPU-Z is not NUMA aware. RAM is only accessible by it's own CPU. As a consequence the OS schedules the task only on the Cores which have direct access to the RAM where the data of the process lies.
You can set NUMA node interleaving to on in the BIOS, this will change the behaviour so...
No, the 950 Pro is the only one.
But for C602 and X79 you can mod the UEFI and add the option rom there.
Instructions for AMI BIOS: [Guide] How to get full NVMe support for all Systems with an AMI UEFI BIOS