Building Cheap Miners : My "Secret" - page 4.

HerstellerBTC

newbie

Activity: 14

Merit: 0

Quote from: JonnikM on July 31, 2019, 05:00:22 AM

Quote from: HerstellerBTC on July 31, 2019, 04:59:43 AM

So you use different miners for different purposes?

Yes, exactly. I have a small farm on ETH, 2 miners on BTC, the rest is altcoin mining

And all of your miners are working? No break downs?

JonnikM

newbie

Activity: 28

Merit: 0

Quote from: HerstellerBTC on July 31, 2019, 04:59:43 AM

So you use different miners for different purposes?

Yes, exactly. I have a small farm on ETH, 2 miners on BTC, the rest is altcoin mining

HerstellerBTC

newbie

Activity: 14

Merit: 0

So you use different miners for different purposes?

JonnikM

newbie

Activity: 28

Merit: 0

Quote from: HerstellerBTC on July 31, 2019, 04:18:49 AM

You personally own such rigs? Could you tell more about the miners and where can I get them

So, basically all miners nowadays are called antiminers, in order to not using GPU miners, antiminers like S17 or S17 pro can really make the difference. I buy all antiminers and complemented equipment on globemining com. I buy there for a long time, yesterday, for example, I have ordered another miner there

HerstellerBTC

newbie

Activity: 14

Merit: 0

You personally own such rigs? Could you tell more about the miners and where can I get them

JonnikM

newbie

Activity: 28

Merit: 0

Quote from: HerstellerBTC on July 31, 2019, 04:09:16 AM

I want to start mining dash or something like this, have you ever tried mining altcoins?

Yes, I have but you need really good rigs to be on top of the game

HerstellerBTC

newbie

Activity: 14

Merit: 0

I want to start mining dash or something like this, have you ever tried mining altcoins?

JonnikM

newbie

Activity: 28

Merit: 0

Quote from: ?? on ??

Hello buddies, I have just registered here to ask about some aspects of altcoin mining. As I have never tried mining maybe you can share some ideas because there is not much available info on the web, thx

Hi, what are you interested in? Because altcoin mining is a huge niche.

Michael20

member

Activity: 118

Merit: 10

You have the system? up and running? Grin

MinersRus

member

Activity: 214

Merit: 24

@SixLitr

Yes there really is something wrong with that last processor and I believe that it is probably memory related.

Your tests are showing that "sudo ./TurionPowerControl -psmax 1" is working properly.

The other three processors go from 1220 H/s per node to 1310 H/s per node or a gain of 7.4%.

As for that last processor is the system reporting any defective memory modules or complaining about non-optimal memory configuration when booting?

You can set in the BIOS under memory settings that a memory test is always run when booting.

Also to check for bad and or mismatched memory you could swap the memory between processors and see if the problem moves to another processor.

You can also use TurionPowerControl to check the DRAM Configuration Status

Code:

sudo ./TurionPowerControl -dram

Example from my quad Opteron 6348 Dell R815:

Code:

DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=63
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=63
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 1 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=61
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=8 TrwtTO=7 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=60
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 2 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=64
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=61
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 3 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=61
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=60
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 4 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=62
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=63
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 5 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=61
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=8 TrwtTO=7 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=61
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 6 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=63
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=63
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Node 7 ---
DCT0: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=9 TrwtTO=8 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=61
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

DCT1: DDR3 frequency: 1333 MHz
Tcl=9 Trcd=9 Trp=9 Tras=24 Access Mode:1T Trtp=5 Trc=33 Twr=10 Trrd=4 Tcwl=7 Tfaw=20
TrwtWB=8 TrwtTO=7 Twtr=5 Twrrd=1 Twrwrsdsc=1 Trdrdsdsc=1 Tref=2 Trfc0=4 Trfc1=2 Trfc2=4 Trfc3=4 MaxRdLatency=59
LDIMM0=EMPTY/EMPTY LDIMM1=OK/OK LDIMM2=EMPTY/EMPTY LDIMM3=EMPTY/EMPTY

Hopefully you can spot the memory problem from this data. MaxRdLatency and DDR3 frequency are probable the most important along with the LDIMM results.

In my system I have four 2GB memory sticks for each Opteron 6348. That means two sticks for each of the two nodes in each Opteron. Thus the LDIMM1=OK/OK shown above.

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------

On the subject of TurionPowerControl I have found more useful commands.

When I am testing I open three shellinthebox command windows to the server.

In the first one I run the actual benchmark. In the second I run nmon with the "c" key when it opens to observe which cores are active.

In the third one I cd into turionpowercontrol and run various TurionPowerControl commands while the benchmark is running.

This command: sudo ./TurionPowerControl -CM

shows the actual Pstate for all cores in real time. If you run this during testing without doing the "TurionPowerControl -psmax 1" you will see that the Pstate flips from PS1 and PS2 all the time. But if you have done "TurionPowerControl -psmax 1" it stays locked in PS1 which is why you get the gain.

To show useful information about Hardware Thermal Control features of the processors in the system
sudo ./TurionPowerControl -htc

The Dell R815 shows 70c as the high temperature limit and that that temp was exceeded before (which throttles the processor).

This command sets the high temperature limit on active node as a threshold for thermal control activity
sudo ./TurionPowerControl -htctemplimit

I did "sudo ./TurionPowerControl -htctemplimit 75" and reran the benchmark but it resulted in no gain in H/s so I left it at the default 70c

SixLitr

jr. member

Activity: 62

Merit: 2

The HT Assist was in fact enabled, and C1E was disabled, swapped those.

Result of TurionPowerControl:

Code:

Main processor is Family 15h (Bulldozer/Interlagos/Valencia) Processor
   Family: 0xf Model: 0x1 Stepping: 0x2
   Extended Family: 0x15 Extended Model: 0x1
   Package Type: 0x3 BrandId: 0x0
Machine has 8 nodes
Processor has 8 cores
Processor has 8 p-states
Processor has 2 boost states

Power States table:
-- Node: 0 Core 0
core 0 pstate 0 (pb0) - En:1 VID:19 FID:16 DID:0.00 Freq:3200 VCore:1.3125
core 0 pstate 1 (pb1) - En:1 VID:33 FID:10 DID:0.00 Freq:2600 VCore:1.1375
core 0 pstate 2 (p0) - En:1 VID:39 FID:7 DID:0.00 Freq:2300 VCore:1.0625
core 0 pstate 3 (p1) - En:1 VID:41 FID:5 DID:0.00 Freq:2100 VCore:1.0375
core 0 pstate 4 (p2) - En:1 VID:44 FID:2 DID:0.00 Freq:1800 VCore:1.0000
core 0 pstate 5 (p3) - En:1 VID:47 FID:0 DID:0.00 Freq:1600 VCore:0.9625
core 0 pstate 6 (p4) - En:1 VID:50 FID:12 DID:1.00 Freq:1400 VCore:0.9250
core 0 pstate 7 (p5) - En:0 VID:0 FID:0 DID:0.00 Freq:1600 VCore:1.5500

I went ahead and changed the power control from 6 to 1 successfully and results for all 4 CPUs are [9073H/s]

Code:

Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1318.61 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1313.64 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1313.27 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1314.09 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1312.53 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1303.8 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 604.595 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 596.604 hashes per second

A performance increase to be sure for the first 3 CPUs, however I suspect there may be some issues with that last CPU, or possibly a memory placement/configuration issue.

Running the test again now at psmax 6 results a slightly higher performance overall, simply because that last CPU isn't reporting so low [9266H/s]

Code:

Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1188.97 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1184.08 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1183.85 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1184.04 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1183.31 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1178.12 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1088.27 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1078.13 hashes per second

MinersRus

member

Activity: 214

Merit: 24

@SixLitr Your tests showing no difference using TurionPowerControl do not make sense.

I have the exact same Dell R815 servers and both the 6200 and 6300 Series Opterons in them gain 10% by applying the "sudo ./TurionPowerControl -psmax 1" command.

TurionPowerControl must be run as root, thus the sudo in front of all TurionPowerControl commands.

Lets try some troubleshooting to see if the issue can be found.

In the BIOS make sure these processor settings are set to the following values:

Code:

HT assist: Disabled
Core Performance Boost Mode: Enabled
Processor HPC mode: Disabled
C1E: Enabled

If one or more of these are set incorrectly then TurionPowerControl may not work correctly or at all.

With that taken care of run this command:

Code:

sudo ./TurionPowerControl -l

and note the results:

from my quad 6348 Opteron system I got:

Code:

Main processor is Family 15h (Bulldozer/Interlagos/Valencia) Processor
   Family: 0xf Model: 0x2 Stepping: 0x0
   Extended Family: 0x15 Extended Model: 0x2
   Package Type: 0x3 BrandId: 0x0
Machine has 8 nodes
Processor has 6 cores
Processor has 8 p-states
Processor has 2 boost states

Power States table:
-- Node: x Core x
core 0 pstate 0 (pb0) - En:1 VID:24 FID:18 DID:0.00 Freq:3400 VCore:1.2500
core 0 pstate 1 (pb1) - En:1 VID:30 FID:15 DID:0.00 Freq:3100 VCore:1.1750
core 0 pstate 2 (p0) - En:1 VID:40 FID:12 DID:0.00 Freq:2800 VCore:1.0500
core 0 pstate 3 (p1) - En:1 VID:43 FID:9 DID:0.00 Freq:2500 VCore:1.0125
core 0 pstate 4 (p2) - En:1 VID:47 FID:5 DID:0.00 Freq:2100 VCore:0.9625
core 0 pstate 5 (p3) - En:1 VID:50 FID:2 DID:0.00 Freq:1800 VCore:0.9250
core 0 pstate 6 (p4) - En:1 VID:54 FID:12 DID:1.00 Freq:1400 VCore:0.8750
core 0 pstate 7 (p5) - En:0 VID:0 FID:0 DID:0.00 Freq:1600 VCore:1.5500


--- Node x:
Processor Maximum PState: 6
Processor Startup PState: 6
Processor Maximum Operating Frequency: 3400 MHz

Note that current "Processor Maximum PState" is 6

Now run:

Code:

sudo ./TurionPowerControl -psmax 1

Command should respond with "Done."

Now again run

Code:

sudo ./TurionPowerControl -l

and make sure that the results at the end show that Processor Maximum PState is now 1:

Code:

--- Node x:
Processor Maximum PState: 1
Processor Startup PState: 6
Processor Maximum Operating Frequency: 3400 MHz

If you get the correct "Processor Maximum PState: 1" then you should gain 10% in hash rates.

SixLitr

jr. member

Activity: 62

Merit: 2

Quote

Does that mean that "HT assist" was already disabled in the BIOS?

I completely forgot to check this honestly - the R&D facility is 30min away so I will have to head back tomorrow and check this setting, most of the time I am just remoting into these servers. Based on the results below I think this may in fact be the culprit to that low hash rate on the last CPU.

Quote

I have discovered that TurionPowerControl binary can be downloaded here:

https://github.com/mh0rst/turionpowercontrol/releases

Just download the tpc-0.44-rc2.tar.gz file (or tpc-0.44-rc2.zip) and unzip it.

In the tpc-0.44-rc2/bin/Ubuntu-amd64 folder you will see the TurionPowerControl binary

I found that yesterday and still didn't make a difference on 18LTS unfortunately.

And now for the latest update: I went ahead and put 16LTS on a USB thumb drive (great idea don't know why I didn't think of it sooner) BUT I accidentally flashed the image to the hard drive array instead of booting off the USB, oh well, that's what these servers are for right now. So now I can compare 16 vs 18 head to head if there are any subtle differences in OS versions, but I digress.

I ran the baseline test again with all 4 CPUs and got slightly better results [9388H/s] - I even went back after and set psmax to 6 again to check and got [9395H/s] so it's fairly consistent and repeatable.

Code:

seq 0 7 | xargs -P 0 -I node numactl -N node ./randomx-benchmark --mine --largePages --jit --nonces 100000 --init 8 --threads 8
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
Memory initialized in 11.2737 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 11.2907 s
Initializing 8 virtual machine(s) ...
Memory initialized in 11.2882 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Running benchmark (100000 nonces) ...
Memory initialized in 11.3152 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 11.3287 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 11.342 s
Initializing 8 virtual machine(s) ...
Memory initialized in 11.3486 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 11.3483 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Running benchmark (100000 nonces) ...
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1220.71 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1219.81 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1218.37 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1217.59 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1217.56 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1216.92 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1043.43 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1038.67 hashes per second

I then ran the same 4 CPU test with TurionPowerControl settings at Processor Maximum PState: 1 and it got worse [8756H/s]

Code:

seq 0 7 | xargs -P 0 -I node numactl -N node ./randomx-benchmark --mine --largePages --jit --nonces 100000 --init 8 --threads 8
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
Memory initialized in 10.6503 s
Initializing 8 virtual machine(s) ...
Memory initialized in 10.6667 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6677 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6707 s
Initializing 8 virtual machine(s) ...
Memory initialized in 10.6735 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Running benchmark (100000 nonces) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6835 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6931 s
Initializing 8 virtual machine(s) ...
Memory initialized in 10.6934 s
Initializing 8 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Running benchmark (100000 nonces) ...
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1310.76 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1309.99 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1310.33 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1309.63 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1305.83 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1305.44 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 454.509 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 454.231 hashes per second

And finally I ran the same 4 CPU test at psmax 1 with only 7 VM's and got this [8612H/s]

Code:

seq 0 7 | xargs -P 0 -I node numactl -N node ./randomx-benchmark --mine --largePages --jit --nonces 100000 --init 8 --threads 7
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
RandomX benchmark v1.0.4
 - full memory mode (2080 MiB)
 - JIT compiled mode
 - hardware AES mode
 - large pages mode
Initializing (8 threads) ...
Memory initialized in 10.6593 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6523 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6576 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6864 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 10.6983 s
Initializing 7 virtual machine(s) ...
Memory initialized in 10.6983 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Running benchmark (100000 nonces) ...
Memory initialized in 13.1069 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Memory initialized in 13.1296 s
Initializing 7 virtual machine(s) ...
Running benchmark (100000 nonces) ...
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1267.87 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1267.82 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1267.58 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1267.42 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1264.88 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 1258.68 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 521.747 hashes per second
Calculated result: d6660144e9a2e68bf47d7cc8afc206672e72f82dfff69fe0d974531e85f7504f
Performance: 501.978 hashes per second

MinersRus

member

Activity: 214

Merit: 24

Quote from: SixLitr on July 16, 2019, 12:58:28 PM

@MinersRus

Unfortunately no, I am unable to improve that last processor but I am still working on things.

Does that mean that "HT assist" was already disabled in the BIOS?

These are the other settings I have in the processor settings:

HT assist: Disabled
Core Performance Boost Mode: Enabled
Processor HPC mode: Disabled
C1E: Enabled

Quote from: SixLitr

No luck getting TurionPowerControl working either.

I have discovered that TurionPowerControl binary can be downloaded here:

https://github.com/mh0rst/turionpowercontrol/releases

Just download the tpc-0.44-rc2.tar.gz file (or tpc-0.44-rc2.zip) and unzip it.

In the tpc-0.44-rc2/bin/Ubuntu-amd64 folder you will see the TurionPowerControl binary

MinersRus

member

Activity: 214

Merit: 24

Quote from: SixLitr on July 16, 2019, 12:58:28 PM

@MinersRus

Unfortunately no, I am unable to improve that last processor but I am still working on things. No luck getting TurionPowerControl working either. Will update if I get anything different (positive or negative). If anyone reading has any ideas on how to get Turion installed on Ubuntu 18 please let me know Wink

- I'd hate to have to downgrade these servers to 16, but if I have to I will to see if anything changes.

For testing without downgrading you can install Ubuntu 16.04 LTS desktop to a USB stick and boot from that stick for testing.

All of my Dell R815's are running off a USB stick.

For testing you can just attach it to the front or back.

For production I have it installed inside at the front left side where there is an internal USB port.

-------------------------------------------------------------------------------------------------------------------

Just a FYI but not all USB sticks are fast or reliable.

I have found that the "SanDisk 16GB Cruzer GLIDE USB Flash Drive" is both fast and reliable.

I buy them new off eBay and just picked up another four of them for $17.50

https://www.ebay.com/itm/Lot-of-4-SanDisk-16GB-Cruzer-GLIDE-USB-Flash-Pen-Drive-2-Lanyards-64GB-total/251883862710

SixLitr

jr. member

Activity: 62

Merit: 2

@MinersRus

Unfortunately no, I am unable to improve that last processor but I am still working on things. No luck getting TurionPowerControl working either. Will update if I get anything different (positive or negative). If anyone reading has any ideas on how to get Turion installed on Ubuntu 18 please let me know Wink

- I'd hate to have to downgrade these servers to 16, but if I have to I will to see if anything changes.

MinersRus

member

Activity: 214

Merit: 24

@SixLitr: Were you able to improve the hash rate for the last processor?

Quote

The results show around 2425 H/s for the first three CPU's but only 1545 H/s for the last CPU.

The probable reason for the reduced result is that "HT assist" is enabled in the BIOS. If "HT assist" is enabled it reduces L3 cache by 2MB.
Check the BIOS settings under Processor settings and turn off "HT assist".

MinersRus

member

Activity: 214

Merit: 24

Quote from: mingdao on July 12, 2019, 02:29:20 AM

Quote from: MinersRus on July 11, 2019, 03:34:51 PM

Quote from: mingdao on July 11, 2019, 06:36:54 AM

Quote from: MinersRus on July 10, 2019, 07:03:49 PM

Quote from: mingdao on July 09, 2019, 05:44:47 PM

Guys,
I am playing around with Proliant G7 4x E7-4870 and I cannot figure out the cpu threads config for XMR-STAK.
Getting either thread 0 error or very low hashrate.

What is your OS (Linux or Windows) version number?

Post your Nvidia.txt config file here so I can look at it and give you some help.

Hi. I booted HiveOS on this machine from the USB drive.

I no longer mine using HiveOS so I suggest going to the HiveOS Forum and asking them for help.
https://forum.hiveos.farm

Ar you using plain Ubuntu?

Yes I am on Ubuntu 16.04

SixLitr

jr. member

Activity: 62

Merit: 2

I have all the updates applied and that includes build-essential. No luck and I am contemplating re-installing with Ubuntu 16LTS just to see if that's the case. TurionPowerControl hasn't been updated in years.

MinersRus

member

Activity: 214

Merit: 24

Quote from: SixLitr on July 15, 2019, 12:40:21 PM

I just double-checked and I had these R815's updated to Ubuntu 18 LTS, that might be why I'm having issues.

I am not on Ubuntu 18 LTS so I can't check that.

However I did notice I was missing the install build-essential step

You may want to do:

sudo apt install build-essential

and see if the make now works.

Code:

sudo apt install ncurses-dev
sudo apt install build-essential

git clone https://github.com/mh0rst/turionpowercontrol.git
cd turionpowercontrol
make

Topic: Building Cheap Miners : My "Secret" - page 4. (Read 60234 times)