Monero Mining Performance
- Thread starter Patrick
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Pretty similar indeed. Thanks! I wonder is power consumption is the same. Should be pretty close.
How did I miss that........tons of used Opterons for cheap on ebay plus some cheap new dell motherboards.......hummmm
What were those Dell C6100's with dual quad opteron nodes? I thought it was a C6105 but this looks like it's the older opteron 4100. You'd also need 6200 or 6300 for AES right?
Dell PowerEdge C6105 W QTY 4 NODES 8 x SIX CORE AMD 4174 HE 2.30GHz SERVER QTY | eBay
Dell PowerEdge C6105 W QTY 4 NODES 8 x SIX CORE AMD 4174 HE 2.30GHz SERVER QTY | eBay
My memory is terrible. C6145 not 6105
Dell Poweredge C6145 Rack Server L5 Chassis MidPlane Board Fan Controller Board | eBay
Dell C6145 4-Socket System Board V3 With Tray 6F0DX | eBay
Dell Poweredge C6145 Rack Server L5 Chassis MidPlane Board Fan Controller Board | eBay
Dell C6145 4-Socket System Board V3 With Tray 6F0DX | eBay
It is a ASUS 4 nodes 2u chassis Asus RS720Q-E8-RS8-P Barebone System - 2U Rack-mountable - Intel C612 Chipset - 4 Number of Node(s) - Socket R3 (LGA2011-3) - 2 x Processor Support - Newegg.com
The entire chassis consumes 640watt = 160watt per node.
The entire chassis consumes 640watt = 160watt per node.
This is likely a (very) bad idea. But Dockerized Monero mining using FreeNAS Corral.
That is right FreeBSD -> bhyve VM -> Docker as a mining platform.
There are a lot of reasons that this is a bad idea, but I think I will make a new image anyway.
That is right FreeBSD -> bhyve VM -> Docker as a mining platform.
There are a lot of reasons that this is a bad idea, but I think I will make a new image anyway.
Sounds like fun. Why not mine directly in a jail?
By the way, do you guys know the power consumption difference (idle or while mining) between an E5 2630L and a 2650L? more than 100H/s more make the 2650 appealing but I'm not sure how much more watts it would need - and TDP isn't of much use for that IMO.
Well, the FreeNAS Corral is pushing Docker heavily, our mining images are in Docker. It is just that Docker on Corral runs in a bhyve VM.
I just added a 250 gig hard drive into my youngest daughters gaming pc, FX 8320 cpu watercooled and overclocked to 4.4 GHz with a GTX 970 video card.
Installing XMR-STAK-NVIDIA and XMR-STAK-CPU as I am typing this.
No she does not know..............LOL
Installing XMR-STAK-NVIDIA and XMR-STAK-CPU as I am typing this.
No she does not know..............LOL
GTX 970 270 H/s
FX 8320 ~405 H/s on 7 cores.
L2 = 4x2mb
L3= 8mb
EDIT: Rebooted and the graphics driver crapped out....urrrr
EDIT 2: It has to be an error with the memlock instructions in the xmr-stack guide. When I installed it to my dual xeon pc it would go straight to the command line and error complaining about x not starting, was an easy fix.
This time it just hangs at a black screen, so reaching for the live cd once again I dive into the guts of Ubuntu.
The instructions tell you to:"
n Linux you will need to configure large page support `sudo sysctl -w vm.nr_hugepages=128` and increase your
ulimit -l. To do do this you need to add following lines to /etc/security/limits.conf:
* soft memlock 262144
* hard memlock 262144"
That caused issues both times I installed XMR-STAK. The first time I just set the min and max to the amount of ram I have and that worked.
I think I will use "
FX 8320 ~405 H/s on 7 cores.
L2 = 4x2mb
L3= 8mb
EDIT: Rebooted and the graphics driver crapped out....urrrr
EDIT 2: It has to be an error with the memlock instructions in the xmr-stack guide. When I installed it to my dual xeon pc it would go straight to the command line and error complaining about x not starting, was an easy fix.
This time it just hangs at a black screen, so reaching for the live cd once again I dive into the guts of Ubuntu.
The instructions tell you to:"
n Linux you will need to configure large page support `sudo sysctl -w vm.nr_hugepages=128` and increase your
ulimit -l. To do do this you need to add following lines to /etc/security/limits.conf:
* soft memlock 262144
* hard memlock 262144"
That caused issues both times I installed XMR-STAK. The first time I just set the min and max to the amount of ram I have and that worked.
I think I will use "
- * soft memlock unlimited
- * hard memlock unlimited
Last edited:
Booting with a Ubuntu 16.04 live disk and editing the etc/security/limits.conf and changing both soft and hard memlock to "unlimited" let me boot.
Ran XMR-STAK-CPU and it still would error out with "**Error: MEMORY ALLOC FAILED: mmap failed**"
So I changed hugepages=128, 128 is what is shown in the readme, to hugepages=256 and that fixed it.
Results:
AMD FX 8320 at 4.4 GHz
XMR-STAK-CPU: only max listed, after a few minutes it would settle on a slightly lower hash rate.
8 threads= 305 H/s
7 threads= 435 H/s
6 threads= 329 H/s
5 threads= 324 H/s
4 threads= 219 H/s
3 threads= 217 H/s
2 threads= 111 H/s
1 thread = 107 H/s
The 8 core FX cpu's have a 8MB L3 cache AND each module pair has 2MB of L2 cache so the cpu has a total of 16 MB of cache.
Looks like the sweet spot is 7 threads but it sure puts out the heat when 7 threads are cranking, I may lower the overclock depending on how hot the cpu gets after running it a while.
EDIT: Ran 7 threads for five minutes and the cpu temp was 56c and it settled down to 425H/s, I think max is 62c so I might downclock it a little and lower the vcore voltage. Fans on the radiator are LOUD!! LOL
EDIT2: Lowered it to 4.2 GHz from 4.4 GHz and voltage to 1.35 v from 1.38v and lowered the voltage offset to .050mv from .10.
Now at 4.2 GHz with 7 threads running with XMR-STAK-CPU :
7 threads = 413 H/s
Cpu temp is 45.2c and fans are much MUCH quieter so thats a good compromise only about ~12-13 H/s slower but much quieter.
So i'm going to fire up XMR-STAK-NVIDIA and see what happens.
EDIT3: I am really impressed with the FX8320 cpu and the FX 8 cores should get CHEAP after all the Ryzen cpu's are out.
Looks like with both the cpu and gpu working the cpu temps are up a bit so I might lower the overclock a little more.
Ran XMR-STAK-CPU and it still would error out with "**Error: MEMORY ALLOC FAILED: mmap failed**"
So I changed hugepages=128, 128 is what is shown in the readme, to hugepages=256 and that fixed it.
Results:
AMD FX 8320 at 4.4 GHz
XMR-STAK-CPU: only max listed, after a few minutes it would settle on a slightly lower hash rate.
8 threads= 305 H/s
7 threads= 435 H/s
6 threads= 329 H/s
5 threads= 324 H/s
4 threads= 219 H/s
3 threads= 217 H/s
2 threads= 111 H/s
1 thread = 107 H/s
The 8 core FX cpu's have a 8MB L3 cache AND each module pair has 2MB of L2 cache so the cpu has a total of 16 MB of cache.
Looks like the sweet spot is 7 threads but it sure puts out the heat when 7 threads are cranking, I may lower the overclock depending on how hot the cpu gets after running it a while.
EDIT: Ran 7 threads for five minutes and the cpu temp was 56c and it settled down to 425H/s, I think max is 62c so I might downclock it a little and lower the vcore voltage. Fans on the radiator are LOUD!! LOL
EDIT2: Lowered it to 4.2 GHz from 4.4 GHz and voltage to 1.35 v from 1.38v and lowered the voltage offset to .050mv from .10.
Now at 4.2 GHz with 7 threads running with XMR-STAK-CPU :
7 threads = 413 H/s
Cpu temp is 45.2c and fans are much MUCH quieter so thats a good compromise only about ~12-13 H/s slower but much quieter.
So i'm going to fire up XMR-STAK-NVIDIA and see what happens.
EDIT3: I am really impressed with the FX8320 cpu and the FX 8 cores should get CHEAP after all the Ryzen cpu's are out.
Looks like with both the cpu and gpu working the cpu temps are up a bit so I might lower the overclock a little more.
Last edited: