Terrible performance compared to Blagominer on large system

[peteraclausen]

I have a large miner with 108 hard drives and a total capacity of 407.5TB. My server consists of a dual E5-2660 V4 CPU server with 64GB of RAM. All 108 drives are directly connected via LSI 9211-8i and 9200-8e HBA's via SAS2 backplane expanders. Using the AVX2 version of the Blago miner, my plots are scanned in about 20 seconds at a speed of about 5,000 MB/s.

Switching to creepMiner, my performance slows to a crawl. I have the following in my mining.conf file:

"bufferChunkCount" : 16, "cpuInstructionSet" : "AVX2", "getMiningInfoInterval" : 3, "gpuDevice" : 0, "gpuPlatform" : 0, "intensity" : 108, "maxBufferSizeMB" : 55000, "maxPlotReaders" : 108, "processorType" : "CPU", "rescanEveryBlock" : false, Upon startup, I do see the following:

Total plots size: 407.51TB Mining Intensity: 108 Max plot readers: 108 Processor type: CPU CPU instruction set: AVX2

creepminer takes about 300 seconds to scan my plots compared to about 20 seconds for Blago miner. So something seems off to say the least.

Here's a view from creepMiner after completing a scan:

http://www.cstone.net/~dk/creepminerscanspeed.PNG

And from Blago miner for comparison:

http://www.cstone.net/~dk/dualargonscan5.PNG

EDIT: for some reason the above links don't work unless you cut and paste them into a new tab in your browser.

Perhaps my hardware is not compatible with creepMiner for some reason?

[amiller2k17]

I am no expert on creepMiner but I recall some recommended settings in the install guides. You should probably review/tweak the settings for your best performance. To me though, your intensity/readers seems a bit high.

Note, your images just give a 404 error.

[peteraclausen]

I never was able to locate any install guides, just the Wiki which explained the items in the config file. I have tried various things in the config file to no avail. I'll just stick to Blago's miner which just works.

As per my original post, so see the images, you have to cut and paste the URLs I posted. Not sure why they don't work directly like on all other forums.

[nitr8]

Hi peteraclausen,

I was also having read issues on a SAN had slow read times ~120s on 50TB LUN's but can read / write at 2000MB/s and it did not telly up.

It all comes down to two things / hardware layout and mining.config setup. Please jump on discord and look me up, lets discuss there then update this once we find you an optimal solution.

[jretropie]

I also report performance issue on Case 311 and never get resolve. Reading this post make more sense because my plot is also large (140TB): https://github.com/Creepsky/creepMiner/issues/311

Notice that CreepMiner read fast when there's 6 or less Hard Drive. When I try add more, the reading speed start to get slower.

In my case with 17 drives, Blago read all plots never less than 100MB/s. CreepMiner read most plots with 60~90MB/s.

6 Hard Disks reading samples:

16.03.2018 09:52:02 (12, C:\projects\creepminer\src\plots\PlotReader.cpp, 345, Information): Dir I:\plots read in 11.451s (~162.66 MB/s) 1 file (7.28 TB) 16.03.2018 09:52:02 (8, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): E:\plots\9187687042291742247_12518030000000_7372800_7372800 (1.76 TB) read in 3.518s (~127.90 MB/s) 16.03.2018 09:52:02 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118235165824_4194304_4194304 (1.00 TB) read in 2.262s (~113.17 MB/s) 16.03.2018 09:52:03 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118115165824_4194304_4194304 (1.00 TB) read in 2.333s (~109.72 MB/s) 16.03.2018 09:52:03 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918025165824_4194304_4194304 (1.00 TB) read in 2.307s (~110.96 MB/s) 16.03.2018 09:52:03 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118239360128_1157448_1157448 (282.58 GB) read in 0.774s (~91.26 MB/s) 16.03.2018 09:52:03 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 345, Information): Dir F:\plots read in 13.024s (~143.01 MB/s) 8 files (7.28 TB) 16.03.2018 09:52:03 (7, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): D:\plots\9187687042291742247_12718040000000_7372800_7372800 (1.76 TB) read in 3.929s (~114.52 MB/s) 16.03.2018 09:52:03 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118119360128_1157448_1157448 (282.58 GB) read in 0.770s (~91.74 MB/s) 16.03.2018 09:52:03 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 345, Information): Dir G:\plots read in 13.256s (~140.51 MB/s) 8 files (7.28 TB) 16.03.2018 09:52:04 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918029360128_1157448_1157448 (282.58 GB) read in 0.812s (~86.94 MB/s) 16.03.2018 09:52:04 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 345, Information): Dir H:\plots read in 13.490s (~138.07 MB/s) 8 files (7.28 TB) 16.03.2018 09:52:04 (7, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): D:\plots\9187687042291742247_12718047372800_1028096_1028096 (251.00 GB) read in 0.731s (~85.83 MB/s) 16.03.2018 09:52:04 (7, C:\projects\creepminer\src\plots\PlotReader.cpp, 345, Information): Dir D:\plots read in 13.980s (~133.24 MB/s) 5 files (7.28 TB) 16.03.2018 09:52:05 (8, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): E:\plots\9187687042291742247_12518040000000_7372800_7372800 (1.76 TB) read in 3.311s (~135.90 MB/s) 16.03.2018 09:52:06 (8, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): E:\plots\9187687042291742247_12518047372800_1028096_1028096 (251.00 GB) read in 0.753s (~83.32 MB/s) 16.03.2018 09:52:06 (8, C:\projects\creepminer\src\plots\PlotReader.cpp, 345, Information): Dir E:\plots read in 15.588s (~119.50 MB/s) 5 files (7.28 TB)

17 Hard Drives reading samples:

2018 20:17:20 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118210000000_4194304_4194304 (1.00 TB) read in 2.029s (~126.15 MB/s) 15.03.2018 20:17:20 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918000000000_4194304_4194304 (1.00 TB) read in 2.320s (~110.34 MB/s) 15.03.2018 20:17:22 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118090000000_4194304_4194304 (1.00 TB) read in 4.276s (~59.86 MB/s) 15.03.2018 20:17:22 (16, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): m:\plots\9187687042291742247_21018010000000_4194304_4194304 (1.00 TB) read in 4.285s (~59.74 MB/s) 15.03.2018 20:17:23 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118214194304_4194304_4194304 (1.00 TB) read in 3.133s (~81.70 MB/s) 15.03.2018 20:17:24 (13, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): J:\plots\9187687042291742247_22118163554432_4194304_4194304 (1.00 TB) read in 5.438s (~47.07 MB/s) 15.03.2018 20:17:24 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918004194304_4194304_4194304 (1.00 TB) read in 3.199s (~80.01 MB/s) 15.03.2018 20:17:25 (7, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): D:\plots\9187687042291742247_12718010000000_7372800_7372800 (1.76 TB) read in 6.502s (~69.21 MB/s) 15.03.2018 20:17:25 (6, C:\projects\creepminer\src\mining\Miner.cpp, 425, Information): 6FK9-PYZN-5G4F-9SB2Z: nonce found (29d 09:46:59) nonce: 38,679,381 in: s:\plots\9187687042291742247_30519296_30519296_30519296 15.03.2018 20:17:25 (6, C:\projects\creepminer\src\network\NonceSubmitter.cpp, 87, Debug): Submit-loop 1 (29d 09:46:59) 15.03.2018 20:17:25 (6, C:\projects\creepminer\src\network\NonceSubmitter.cpp, 109, Information): 6FK9-PYZN-5G4F-9SB2Z: nonce submitted (29d 09:46:59) nonce: 38,679,381 in: s:\plots\9187687042291742247_30519296_30519296_30519296 15.03.2018 20:17:25 (16, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): m:\plots\9187687042291742247_21018014194304_4194304_4194304 (1.00 TB) read in 2.989s (~85.62 MB/s) 15.03.2018 20:17:25 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118094194304_4194304_4194304 (1.00 TB) read in 3.018s (~84.82 MB/s) 15.03.2018 20:17:25 (8, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): E:\plots\9187687042291742247_12518010000000_7372800_7372800 (1.76 TB) read in 7.312s (~61.54 MB/s) 15.03.2018 20:17:26 (14, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): k:\plots\9187687042291742247_20118010000000_7372800_7372800 (1.76 TB) read in 7.489s (~60.08 MB/s) 15.03.2018 20:17:26 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918008388608_4194304_4194304 (1.00 TB) read in 2.729s (~93.80 MB/s) 15.03.2018 20:17:27 (13, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): J:\plots\9187687042291742247_22118167748736_4194304_4194304 (1.00 TB) read in 3.221s (~79.46 MB/s) 15.03.2018 20:17:27 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118218388608_4194304_4194304 (1.00 TB) read in 3.496s (~73.22 MB/s) 15.03.2018 20:17:28 (6, C:\projects\creepminer\src\network\NonceSubmitter.cpp, 124, Debug): JSON confirmation (29d 09:46:59) {"deadline":2540819,"result":"success"} 15.03.2018 20:17:28 (6, C:\projects\creepminer\src\network\NonceSubmitter.cpp, 182, Information): 6FK9-PYZN-5G4F-9SB2Z: nonce confirmed (29d 09:46:59) nonce: 38,679,381 in: s:\plots\9187687042291742247_30519296_30519296_30519296 15.03.2018 20:17:29 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118098388608_4194304_4194304 (1.00 TB) read in 3.179s (~80.52 MB/s) 15.03.2018 20:17:29 (16, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): m:\plots\9187687042291742247_21018018388608_4194304_4194304 (1.00 TB) read in 3.493s (~73.28 MB/s) 15.03.2018 20:17:30 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918012582912_4194304_4194304 (1.00 TB) read in 3.491s (~73.33 MB/s) 15.03.2018 20:17:31 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118222582912_4194304_4194304 (1.00 TB) read in 3.727s (~68.67 MB/s) 15.03.2018 20:17:31 (13, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): J:\plots\9187687042291742247_22118171943040_4194304_4194304 (1.00 TB) read in 3.735s (~68.52 MB/s) 15.03.2018 20:17:32 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118102582912_4194304_4194304 (1.00 TB) read in 3.256s (~78.61 MB/s) 15.03.2018 20:17:32 (7, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): D:\plots\9187687042291742247_12718020000000_7372800_7372800 (1.76 TB) read in 7.305s (~61.60 MB/s) 15.03.2018 20:17:32 (16, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): m:\plots\9187687042291742247_21018022582912_4194304_4194304 (1.00 TB) read in 3.278s (~78.08 MB/s) 15.03.2018 20:17:32 (14, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): k:\plots\9187687042291742247_20118020000000_7372800_7372800 (1.76 TB) read in 6.713s (~67.03 MB/s) 15.03.2018 20:17:33 (8, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): E:\plots\9187687042291742247_12518020000000_7372800_7372800 (1.76 TB) read in 7.104s (~63.34 MB/s) 15.03.2018 20:17:33 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918016777216_4194304_4194304 (1.00 TB) read in 3.113s (~82.23 MB/s) 15.03.2018 20:17:34 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118226777216_4194304_4194304 (1.00 TB) read in 3.437s (~74.48 MB/s) 15.03.2018 20:17:34 (13, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): J:\plots\9187687042291742247_22118176137344_4194304_4194304 (1.00 TB) read in 3.898s (~65.67 MB/s) 15.03.2018 20:17:35 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118106777216_4194304_4194304 (1.00 TB) read in 3.245s (~78.88 MB/s) 15.03.2018 20:17:35 (16, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): m:\plots\9187687042291742247_21018026777216_4194304_4194304 (1.00 TB) read in 3.092s (~82.79 MB/s) 15.03.2018 20:17:36 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918020971520_4194304_4194304 (1.00 TB) read in 3.085s (~82.98 MB/s) 15.03.2018 20:17:37 (9, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): F:\plots\9187687042291742247_13118230971520_4194304_4194304 (1.00 TB) read in 3.360s (~76.17 MB/s) 15.03.2018 20:17:38 (16, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): m:\plots\9187687042291742247_21018030971520_4194304_4194304 (1.00 TB) read in 2.977s (~85.97 MB/s) 15.03.2018 20:17:38 (10, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): G:\plots\9187687042291742247_13118110971520_4194304_4194304 (1.00 TB) read in 3.166s (~80.85 MB/s) 15.03.2018 20:17:39 (13, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): J:\plots\9187687042291742247_22118180331648_4194304_4194304 (1.00 TB) read in 4.183s (~61.20 MB/s) 15.03.2018 20:17:39 (14, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): k:\plots\9187687042291742247_20118030000000_7372800_7372800 (1.76 TB) read in 6.766s (~66.50 MB/s) 15.03.2018 20:17:39 (7, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): D:\plots\9187687042291742247_12718030000000_7372800_7372800 (1.76 TB) read in 7.404s (~60.77 MB/s) 15.03.2018 20:17:39 (11, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): H:\plots\9187687042291742247_21918025165824_4194304_4194304 (1.00 TB) read in 3.345s (~76.51 MB/s) 15.03.2018 20:17:40 (23, C:\projects\creepminer\src\plots\PlotReader.cpp, 293, Information): z:\plots\9187687042291742247_22218000000000_30519296_30519296 (7.28 TB) read in 21.564s (~86.38 MB/s)

[ChrOst]

set "bufferChunkCount" to 4096 -> 4K reads perform way better then 16byte reads.

[jretropie]

It takes forever to read if I set "bufferChunkCount" to 4096

[nitr8]

hey guys,

@ChrOst and @jretropie actually @all could you please enable the benchmark feature and post the benchmark.csv for review?

[jretropie]

benchmark.csv.txt

[ChrOst]

My findings were:

Disks	Memory	Nice	Reader	Intensity	MB/s	BufferChunkSize	CPUMode	Time in s
14	8192	-19	14	14	757,9	4096	performance	43
14	8192	-19	14	14	740,6	8192	performance	44
14	10240	-19	14	14	749,2	8192	performance	43,5
14	10240	-19	14	14	749,2	4096	performance	43,5
14	6144	-19	14	14	740,6	8192	performance	44
14	6144	-19	14	14	749,2	4096	performance	43,5

[jretropie]

I am not sure what ChrOst trying to come up with your findings....... 1) All the reading are slow. 2) All the results are pretty close no matter what's your setting.

Notice that when I set CPU to AVX or AVX2, it made no different on performance. From my experience with Blago miner, Xplotter, and TurboPlotter, performance should pretty much double when switching from AVX to AVX2.

[loppefaaret]

that just means its not a processing blottleneck, but getting the data to the processor bottleneck. try setting intensity 1, and AVX, my round time is 23.5 setting AVX2, cuts the same round time down to 14.5 not a doubling(or halving), but its clearly affecting processing time

one thing i can add to the whole blago vs creep debacle: the way the 2 miners read data - if a disk is otherwise saturated - when a new round starts, Blago will be much more aggressive or better at getting data from a plotfile, vs creepminer: if im plotting to a drive, while mining the same drive, blago will finish reading multiple times quicker than a standard configured creepminer - i have not tested this with a much higher bufferchunkcount as mentioned earlier in this thread - as i was under the impression that it would split the maxbuffersize into bufferchunks according to the wiki

[ChrOst]

@jretropie they are as fast as they can get using the CPU. using BufferChunkSize 16 decreases it by 100MB/s, setting it to 8 decreases it to 540MB/s. Actually I should have put those values in aswell.

Currently struggeling with a CUDA Card.

[jretropie]

@ChrOst I am not sure what you trying to do here but 757MB/s is VERY slow if you have 14 Disks. Consider Blago reads every disk more than 100MB/s (100 x 14). You should get at least 1400MB/s total.

[ChrOst]

@jretropie yes. it can read over 1600MB/s. BUT: the CPU won't do the Job during mining. As it has to calculate some stuff after reading. And that Xeon just can do about 800MB/s. My current Bottleneck is the Verfication Proccess. Thats why I want to use a Graphicscard.

[T5-R]

Interesting stuff. This is my experience:

On non avx capable cpu's I have found creepminer to be much faster than blago - over 1/3rd if I remember correctly. This was on an AMD Phenom 6core (X6 1090T). I switched to a new Ryzen 6core (r5 1600) some time ago and just continued using creep. Reading this I tried blago again and was surprised that it is indeed much faster. ~98TB blago: 20s (read speed around 1.5GB/s) creep: 30-35s (read speed around 0.9GB/s) (I found that a bufferchunkcount of around 1000 works best for me. read 12-16 threads, intensity 6-9) In Taskmanager one thing I noticed is that creep uses a lot more kernel cpu time than blago.

@ChrOst I don't think the bufferchunkcount works like you think - it is the number of buffers, not the size. It divides the whole buffer in x parts.

[JohnnyFFM]

Hey, just to report I have the same issue. Could be windows related, as I only see win guys complaining. I run 20 disks, blago ~27s, creepminer ~52s, roughly double the time. SMR drives. cpu is i7-7700 using avx2. I played around with all params, no luck. Neither my CPU nor my HDD I/O is maxed will running creep. Hyperthreadding issue? I/O issue? no idea. back to blago for now.

[T5-R]

Here some perfmon capture I took. I marked blago red and creep green. %Interrupt and %Privileged time is highlighted. We can clearly see that creepminer generates a lot more interrupts when reading the plots (up to 50% cpu usage). I don't now the reason, but to me it seems like the "slow performance" is definitely coming from the way the plots are being read.

DataCollector01.zip

[sgabay242]

I would like to report the same issue, our build runs with Ubuntu 16.04 using CreepMiner latest with NVIDIA Cuda on Titan XP, 200TB scans are well over 1 minute consistently.

When we switch to JMiner, scans on the same plot same hardware is performed in 27 to 30 seconds

Same hardware on Windows 10 x64 with Blago miner on AVX2 is about 22 to 24 seconds

Something worth to mention, Creepminer assumes and report the Plot size to be around 190TB, but when using Jminer it reports 204TB on the same hardware, so for whatever reason the size detected is inconsistent 14TB delta is HUGE!

Anybody have done any performance test with Burst to scan using EXT4 vs BTRFS, vs XFS, how about switching the Linux IO scheduler running on noop, anticipatory, deadline, cfq

Thanks! Saul

[T5-R]

@sgabay242 I can't confirm that creepminer reports wrong plotsize. The calculation seems to be different between blago and creep (maybe GiB vs GB or something) but otherwise it was correct for me. Maybe you have some overlap or damaged/missing plots.

[chrisf4lc0n]

@sgabay242 I can confirm the same finding, much longer scan times vs jMiner even o a small plot of 17TB. jMiner completes it in just over 9 seconds when creepMiner requires 20 seconds to complete it. 2 GPUs have been tested R7 250 and RX 480 (I actually get the same times on both GPUs of 9 seconds on my small plot with jMiner).

[brmmm3]

1.9.0.7 compared to 1.9.0.5 seems to have a performance problem too. I've created a fork with BFS (Burst File System, not really a file system, but just a table in the first 1024 bytes which describes where to find which plot data on disk) support. BFS is for raw disk access without any file system overhead -> more speed and more space. 1.9.0.5 works relatively well with 15% higher read speed (disk is connected via USB 3.0. Possible performance improvemend, measured with dd, is +32%). But 1.9.0.7 doesn't show this performance improvement. The forked version can be found here: https://github.com/brmmm3/creepMiner/tree/development A tool for creating and managing a BFS disk (bfs.py) can be found here: https://github.com/brmmm3/plotTools Would be great to see BFS support in the official branch.

[Creepsky]

Thank you everybody for reporting. The problem is file caching. Especially Windows is using an "intelligent" algorithm to pre-fetch parts of a file while reading it. Blago miner is using uncached file streams, creepMiner (and I think jMiner too) is using the implementation of the C(++) standard lib. In 1.9.0.5 (like @brmmm3 said) I implemented these low level file streams, but reverted the change because of instability. However they came back in 1.9.0.8 (or .9, cant remember). Uncached file streams are very low level, you have to read sectors of the hdd instead of single bytes - so its a lot more error prone.

If you are brave enough, you can test the current pre-compiled dev version - thanks in advance for every bug report.

Please bear in mind, that the whole read/calculate algorithm changed - so most likely you also have to change your config. Most significant options are: maxBufferSizeMB and bufferchunkcount.

[chrisf4lc0n]

Just to quickly report back. creepMiner 1.9.0.14 with OPENCL Scan times are even better than jMiner, with the highest scan speeds ~230MB/s on 2x4TB striped disks!

[T5-R]

I'm in the process of replotting for poc2... so I can't compare the perfomance directly right now But I gave the 1.9.0.15 a try (cpu only) and it seems very fast. I get some errors with this dev version on deadline submission (that it exceeds pool deadline while the config is correct). It is a bummer that you have to do this low-level work.... but the result seems to be very good.

[chrisf4lc0n]

The 14 version seems to have the same problem with wrong deadlines.

[ChrOst]

"poc2StartBlock" <-- what does this say in your mining.conf? change it to 502000 please

[chrisf4lc0n]

It is 502000.

[ChrOst]

clinfo |grep "Global memory size"

vs "maxBufferSizeMB" ? make sure you are using 1900 with a 2GB GPU ( a little less then maximum)

[chrisf4lc0n]

@ChrOst "cpuInstructionSet" : "AVX" "maxBufferSizeMB" 1024 " bufferchunkcount" 4 "processorType" : "CPU"

I think I have played with all the possible options I can think of and I am still getting the error: error-code: 1008 error-text: deadline exceeds deadline limit of the pool

[chrisf4lc0n]

I got to the bottom of the error-code: 1008 error-text: deadline exceeds deadline limit of the pool The 14 as well as 15 versions do not seem to convert POC2 to 1, hence the deadline error, that is regardless the "poc2StartBlock" value. Once POC1 plots only are being used the error does not pop up.

[T5-R]

As this is not related to the original performance issue I guess we shouldn't discuss this here on and on again. The Performance seemed to be fixed in the dev build.

[T5-R]

I tried 1.9.0.16 today (most of my drives are now done with poc2) Noticed that it has strange issues with not finding deadlines or read errors depending on the bufferChunkCount I set. Powers of 2 seem to work. Performance seems quite good however, I'm getting around 1.5GB/s and sub 20s times with 80TB right now (AVX2 R5 1600).

[T5-R]

Just an update if you are interested. I have been running the dev build 1.9.0.16 for a few days now and so far no issues. It runs fine, scan times are around 18sec (a few seconds faster than blago) for my 100TB.