bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 27, 2017, 08:50:00 PM |
|
In the mean time, the other thing that no one has reported is what happens when you run two copies of BBP against two different datadirectories on the Ryzen? If anyone can test that out and report back that would be immensely helpful also. I think inblue was going to try that on a server at one point.
Someone tried that on a Xeon before: I just run 8 times biblepayd with genproclimit=1 on my Xeon E3-1245 v5 @ 3.5GHz and now is 70583.5 hashps Before with one biblepayd with genproclimit=8 I had about 68000 hashps so it's not big diffrent I run gentoo linux 4.7.10-hardened with ZFS filesystem.
But I couldn't find a way to run multiple daemons. The .biblepaycore directory is locked when one daemon is running and I don't know how to point a second daemon to a different directory (or any other way to achieve it). How to do it without editing the source code and recompiling for every instance? Oh wow, great info, I didnt see that post, cool. You can specify the -datadir for each launched instance to be different and share the same box with the same codebase.
|
|
|
|
inblue
|
|
September 27, 2017, 09:51:51 PM Last edit: September 28, 2017, 05:23:17 AM by inblue |
|
In the mean time, the other thing that no one has reported is what happens when you run two copies of BBP against two different datadirectories on the Ryzen? If anyone can test that out and report back that would be immensely helpful also. I think inblue was going to try that on a server at one point.
Someone tried that on a Xeon before: I just run 8 times biblepayd with genproclimit=1 on my Xeon E3-1245 v5 @ 3.5GHz and now is 70583.5 hashps Before with one biblepayd with genproclimit=8 I had about 68000 hashps so it's not big diffrent I run gentoo linux 4.7.10-hardened with ZFS filesystem.
But I couldn't find a way to run multiple daemons. The .biblepaycore directory is locked when one daemon is running and I don't know how to point a second daemon to a different directory (or any other way to achieve it). How to do it without editing the source code and recompiling for every instance? Oh wow, great info, I didnt see that post, cool. You can specify the -datadir for each launched instance to be different and share the same box with the same codebase. OK, thanks! So here's what I did, for documentation purposes: $ cd ~ $ cp -r .biblepaycore/ .biblepaycore2/ $ cd /usr/local/bin $ ./biblepayd -daemon $ ./biblepayd -daemon -datadir=/home/inblue/.biblepaycore2
First I got this error: 2017-09-27 21:19:30 Binding RPC on address ::1 port 39000 failed. 2017-09-27 21:19:30 Binding RPC on address 127.0.0.1 port 39000 failed. 2017-09-27 21:19:30 Unable to bind any endpoint for RPC server 2017-09-27 21:19:30 Error: Unable to start HTTP server. See debug log for details. Then I added this to biblepay.conf in .biblepaycore2: Then after starting it again I got this error: Error: Unable to bind to 0.0.0.0:40000 on this computer. Biblepay Core is probably already running. Error: Failed to listen on any port. Use -listen=0 if you want this. So then I run: $ ./biblepayd -daemon -datadir=/home/inblue/.biblepaycore2 -listen=0 I don't know if it can run normally with listen=0, but it seems to start fine. Now I see both processes and they seem to be running as they should, each consuming about half of the CPU (around 500% each of a 10 core CPU which is 1000%): PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 4447 inblue 20 0 2009596 70020 20180 S 503.7 0.8 24:04.27 biblepayd 4568 inblue 20 0 1861720 68544 20216 S 494.4 0.7 4:43.99 biblepayd But I have no clue how to access separate CLIs for each daemon. When I type ./biblepay-cli getmininginfo, I only get info from one of them. How to get info from the other one? Also, when I type ./biblepay-cli stop, both of them stop.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 01:51:06 AM |
|
Error: Unable to bind to 0.0.0.0:40000 on this computer. Biblepay Core is probably already running. Error: Failed to listen on any port. Use -listen=0 if you want this.
Wild, good deal that you got that far. I havent done it with biblepay yet, but here is a link where someone connects to his second instance of multichain-cli: https://www.multichain.com/qa/2817/how-to-run-datadir-in-multichainIt appears the biblepay-cli command authenticates you through a default rpc port and that guy is specifying a second one
|
|
|
|
cyberpitstop
|
|
September 28, 2017, 01:56:21 AM |
|
I found this Bible Pay just browsing around I didn't even know it existed. Are there any exchange's it's listed on??
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:12:31 AM |
|
Yikes, 250738574 is back! Lots of hashpower pointed at the pool. Last subsidy: 1777 BBP:
250738574 9607 13749.0000 1777.1636 9/27/2017 9:04:05 PM 1007822.08 0.0018179076054448 9/27/2017 9:03:59 PM 121: 6473 (1007822) 122: 6629 (1007822) 123: 6623 (1007822) 124: 6510 (1007822) 125: 6393 (1007822) 126: 6438 (1007822) 127: 6493 (1007822) 128: 6626 (1007822) 129: 6675 (1007822) 130: 6690 (1007822) 131: 6808 (1007822) 132: 6808 (1007822) 133: 6823 (1007822) 134: 6693 (1007822) 135: 6826 (1007822) 136: 6689 (1007822) 137: 6661 (1007822) 138: 6675 (1007822) 139: 6768 (1007822) 250738574: 5463 (1007822) 250738574-1: 7025 (1007822) 250738574-10: 6017 (1007822) 250738574-11: 5877 (1007822) 250738574-12: 5878 (1007822) 250738574-13: 5877 (1007822) 250738574-14: 5857 (1007822) 250738574-2: 7271 (1007822) 250738574-3: 7160 (1007822) 250738574-4: 7101 (1007822) 250738574-5: 6685 (1007822) 250738574-6: 6004 (1007822) 250738574-7: 6708 (1007822) 250738574-8: 6651 (1007822) 250738574-9: 6929 (1007822) 5136-1: 6008 (1007822) 5136-10: 5988 (1007822) 5136-11: 6008 (1007822) 5136-12: 5999 (1007822) 5136-13: 5493 (1007822) 5136-14: 5500 (1007822) 5136-15: 5461 (1007822) 5136-16: 5452 (1007822) 5136-17: 5483 (1007822) 5136-18: 5501 (1007822) 5136-19: 5453 (1007822) 5136-2: 5972 (1007822) 5136-20: 6413 (1007822) 5136-21: 6381 (1007822) 5136-22: 6473 (1007822) 5136-23: 6368 (1007822) 5136-24: 6477 (1007822) 5136-25: 6424 (1007822) 5136-26: 6470 (1007822) 5136-27: 6418 (1007822) 5136-28: 6412 (1007822) 5136-29: 6424 (1007822) 5136-3: 5950 (1007822) 5136-30: 4810 (1007822) 5136-31: 4803 (1007822) 5136-32: 5044 (1007822) 5136-33: 4799 (1007822) 5136-34: 4044 (1007822) 5136-35: 5077 (1007822) 5136-36: 4794 (1007822) 5136-37: 4819 (1007822) 5136-38: 5041 (1007822) 5136-39: 5065 (1007822) 5136-4: 5983 (1007822) 5136-40: 5065 (1007822) 5136-41: 5073 (1007822) 5136-42: 4788 (1007822) 5136-43: 4805 (1007822) 5136-44: 5055 (1007822) 5136-45: 4789 (1007822) 5136-5: 5985 (1007822) 5136-6: 5986 (1007822) 5136-7: 5933 (1007822) 5136-8: 5971 (1007822) 5136-9: 5990 (1007822) 8451: 6995 (1007822) 8452: 6003 (1007822) 8453: 6787 (1007822) 8456: 6685 (1007822) 8457: 6636 (1007822) 8458: 6679 (1007822) 8459: 6064 (1007822) 8460: 6016 (1007822) 8461: 6066 (1007822) 8462: 6005 (1007822) 8463: 6066 (1007822) 8464: 6069 (1007822) 8465: 6006 (1007822) 8466: 6009 (1007822) 8467: 6021 (1007822) 8468: 6049 (1007822) 8469: 6077 (1007822)
I'm going to add a feature tomorrow to sort the block_distribution by height descending,subsidy descending - its in Test now.
Stay ready for the rapture.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:13:20 AM |
|
I found this Bible Pay just browsing around I didn't even know it existed. Are there any exchange's it's listed on??
We're on c-cex so far.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:16:55 AM |
|
I found this Bible Pay just browsing around I didn't even know it existed. Are there any exchange's it's listed on??
We're on c-cex so far. If anyone is just browsing around, wondering what we do, we sponsor orphans from 10% of the blockchain, check this site out: http://accountability.biblepay.org/
Click on Orphans | Orphan Active Sponsored Orphans list. We have about 94 orphans already sponsored.
|
|
|
|
foxy
|
|
September 28, 2017, 05:41:01 AM |
|
I found this Bible Pay just browsing around I didn't even know it existed. Are there any exchange's it's listed on??
Yes it's already listed on exchange you can check it out here: https://coinmarketcap.com/currencies/biblepay/#marketsI wish the price of BBP rise a bit though.
|
|
|
|
wanispwnz
|
|
September 28, 2017, 11:01:07 AM |
|
I found this Bible Pay just browsing around I didn't even know it existed. Are there any exchange's it's listed on??
Yes it's already listed on exchange you can check it out here: https://coinmarketcap.com/currencies/biblepay/#marketsI wish the price of BBP rise a bit though. I hope so too but I think it can take some time.
|
|
|
|
inblue
|
|
September 28, 2017, 02:07:50 PM |
|
Wild, good deal that you got that far. I havent done it with biblepay yet, but here is a link where someone connects to his second instance of multichain-cli: https://www.multichain.com/qa/2817/how-to-run-datadir-in-multichainIt appears the biblepay-cli command authenticates you through a default rpc port and that guy is specifying a second one Thanks for the link. So I accessed the second CLI like this: ./biblepay-cli -datadir=/home/inblue/.biblepaycore2 getmininginfo (I put the listen=0 in the conf file) I actually already tried the above command before and it didn't work, but it's because I was too tired and I typed ./biblepay-cli getmininginfo -datadir=/home/inblue/.biblepaycore2 (so the order of arguments is important). RESULTSAll values are taken after about half an hour of mining. Type of computer: 10-CPU Xeon Skylake. Note that for the pool to register more than one instance, every instance has to be a different worker.Running 1 biblepayd at genproclimit 20: HPS: 52k HPS2: 29kRunning 2 biblepayd at genproclimit 10 each: HPS: 29k + 28k = 57k HPS2: 30k + 30k = 60kRunning 4 biblepayd at genproclimit 5 each: HPS: 15k + 14.5k + 14.5k + 14.5k = 58.5k HPS2: 19k + 22k + 19k + 20k = 80kRunning 10 biblepayd at genproclimit 2 each: HPS: 6k x 10 = 60k HPS2: 12k x 10 = 120kSo the conclusion is obvious - the HPS is about the same regardless of the number of daemons, but HPS2 rises considerably.
The next test I did was running 10 daemons on 10 separate 1-CPU machines (the CPU is the same as in the 10-CPU machine) and also genproclimit 2 as before. This scenario gives: HPS: 10k x 10 = 100k HPS2: 17k x 10 = 170kSo this is even more than if you run 10 daemons on the same machine, which is 60k HPS and 120k HPS2, as noted above. SummarySolo miningOne 10-CPU machine with 1 miner gives 52k HPS. One 10-CPU machine with 10 miners gives 60k HPS. Ten 1-CPU machines with 1 miner each give 100k HPS.Pool miningOne 10-CPU machine with 1 miner gives 29k HPS2. One 10-CPU machine with 10 miners gives 120k HPS2. Ten 1-CPU machines with 1 miner each give 170k HPS2.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:26:17 PM |
|
Wild, good deal that you got that far. I havent done it with biblepay yet, but here is a link where someone connects to his second instance of multichain-cli: https://www.multichain.com/qa/2817/how-to-run-datadir-in-multichainIt appears the biblepay-cli command authenticates you through a default rpc port and that guy is specifying a second one Thanks for the link. So I accessed the second CLI like this: ./biblepay-cli -datadir=/home/inblue/.biblepaycore2 getmininginfo (I put the listen=0 in the conf file) I actually already tried the above command before and it didn't work, but it's because I was too tired and I typed ./biblepay-cli getmininginfo -datadir=/home/inblue/.biblepaycore2 (so the order of arguments is important). RESULTSAll values are taken after about half an hour of mining. Type of computer: 10-CPU Xeon Skylake. Note that for the pool to register more than one instance, every instance has to be a different worker.Running 1 biblepayd at genproclimit 20: HPS: 52k HPS2: 29kRunning 2 biblepayd at genproclimit 10 each: HPS: 29k + 28k = 57k HPS2: 30k + 30k = 60kRunning 4 biblepayd at genproclimit 5 each: HPS: 15k + 14.5k + 14.5k + 14.5k = 58.5k HPS2: 19k + 22k + 19k + 20k = 80kRunning 10 biblepayd at genproclimit 2 each: HPS: 6k x 10 = 60k HPS2: 12k x 10 = 120kSo the conclusion is obvious - the HPS is about the same regardless of the number of daemons, but HPS2 rises considerably.
The next test I did was running 10 daemons on 10 separate 1-CPU machines (the CPU is the same as in the 10-CPU machine) and also genproclimit 2 as before. This scenario gives: HPS: 10k x 10 = 100k HPS2: 17k x 10 = 170kSo this is even more than if you run 10 daemons on the same machine, which is 60k HPS and 120k HPS2, as noted above. SummarySolo miningOne 10-CPU machine with 1 miner gives 52k HPS. One 10-CPU machine with 10 miners gives 60k HPS. Ten 1-CPU machines with 1 miner each give 100k HPS.Pool miningOne 10-CPU machine with 1 miner gives 29k HPS2. One 10-CPU machine with 10 miners gives 120k HPS2. Ten 1-CPU machines with 1 miner each give 170k HPS2.Thank you so much for all that. Let me archive this on our wikipedia page. Wiki Link: http://wiki.biblepay.org/Multiple_Instances
|
|
|
|
inblue
|
|
September 28, 2017, 02:31:15 PM |
|
Thank you so much for all that. Let me archive this on our wikipedia page. You're welcome. You could also add the instructions there on how to run multiple nodes, from my previous posts. I am also hoping for your rough opinion of the results.
|
|
|
|
seeksilence
|
|
September 28, 2017, 02:34:28 PM |
|
So we have to run multiple daemons to get better rewards (HPS2)? It seems the pool reward algorithm is wrong? All the shares are equal no matter the difficulty of the share is?
One 10-CPU machine with 1 miner gives 29k HPS2. One 10-CPU machine with 10 miners gives 120k HPS2.
29k to 120k. it is 4 times reward. This will let us create many many workers in the pool.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:35:26 PM |
|
Thank you so much for all that. Let me archive this on our wikipedia page. You're welcome. You could also add the instructions there on how to run multiple nodes, from my previous posts. I am also hoping for your rough opinion of the results. Thanks. Yeah, I wrote one line about it at the top. Anyone, feel free to edit the page to make it nicer. I am in the middle of starting the slack team guys so I have to stay focused. Im also writing code for the governance web interface for support for the upcoming changes. (We have to convert proposals to hex, have pages to enter proposals, convert to RPC commands, create transactions, send escrows, list the proposals, submit the proposals, etc, and all that is just for the first sanctuary command, etc). Thats why it is important for us to start an IT department over the next 3 days.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:39:37 PM |
|
So we have to run multiple daemons to get better rewards (HPS2)? It seems the pool reward algorithm is wrong? All the shares are equal no matter the difficulty of the share is?
No, the assumption isnt that its wrong, the assumption is the biblehash algorithm favors small rigs. Remember the biblehash algorithm isnt written to reward the fastest proc, its written to reward who solves the biblehash chained verses with a solution- and that requires a full node- take a look at the code for the hash algorithm- it requires you to look at the block, get things every so often when the timestamp needs changed, chain the verses together and execute some business logic from the full node to hash it. Now as far as how do we explain why the pool seems to favor distributed nodes over super machines, I have no clue. Ive looked at the code for the solution reward, and its like this: I hand out 20 equal easy problems to the team, and as you solve those equal easy problems you get a share. Thats how it works. Why multiple machines solve these equal weighted shares 'easier' is still beyond me.
|
|
|
|
nitinum
Member
Offline
Activity: 81
Merit: 10
|
|
September 28, 2017, 02:42:56 PM |
|
WTF?!
|
|
|
|
seeksilence
|
|
September 28, 2017, 02:44:51 PM |
|
Maybe the fast server solve the problems too fast and need to wait for another task. The waiting time is too long for fast server. So, if I have a server, I will create many workers to get better rewards and that could put a burden on the pool.
|
|
|
|
bible_pay (OP)
Full Member
Offline
Activity: 1176
Merit: 215
Jesus is the King of Kings and Lord of Lords
|
|
September 28, 2017, 02:46:21 PM |
|
So we have to run multiple daemons to get better rewards (HPS2)? It seems the pool reward algorithm is wrong? All the shares are equal no matter the difficulty of the share is?
One 10-CPU machine with 1 miner gives 29k HPS2. One 10-CPU machine with 10 miners gives 120k HPS2.
29k to 120k. it is 4 times reward. This will let us create many many workers in the pool.
Btw, you can consolidate your workers so we dont have massive long list of miners in the leaderboard- once you know your avg HPS2, you can point two or more daemons at one workerid and should see the increased HPS2 now. The HPS itself will be wrong, but HPS2 should rise.
|
|
|
|
seeksilence
|
|
September 28, 2017, 02:47:43 PM |
|
OK, I will try this way to see if the hps2 increase. Thanks.
|
|
|
|
inblue
|
|
September 28, 2017, 02:52:13 PM |
|
So we have to run multiple daemons to get better rewards (HPS2)? It seems the pool reward algorithm is wrong? All the shares are equal no matter the difficulty of the share is?
One 10-CPU machine with 1 miner gives 29k HPS2. One 10-CPU machine with 10 miners gives 120k HPS2.
29k to 120k. it is 4 times reward. This will let us create many many workers in the pool.
Btw, you can consolidate your workers so we dont have massive long list of miners in the leaderboard- once you know your avg HPS2, you can point two or more daemons at one workerid and should see the increased HPS2 now. The HPS itself will be wrong, but HPS2 should rise. Ah, I've only looked at HPS when I tested that and I guess I didn't give enough time for HPS2 to rise. But does this mean I can point all of my separate machines to just one workerid, or does that apply only to the nodes on the same machine?
|
|
|
|
|