Yeah I wasn't saying there were insurmountable just pointing out potential issues.
Someone should ask Meni of any adverse affects of a very long "n" in PPLNS. That is essentially what we are doing. I don't know of any conventional pool which has an n that large. Then again I don't know of any conventional pool with an n of > 1 and currently p2pool has an n of ~4. There is pretty much nobody who knows more about mining reward systems then Meni.
Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 746. (Read 2591916 times)
March 16, 2012, 09:13:32 AM
March 16, 2012, 09:00:52 AM
A "hard fork" is, practically, just an update to a new version of p2pool. So most miners wont see any problem in that.
And, if we all switch to the new version/chain/fork at the same time, we wouldnt receive less per found block, since we all begin at zero.
New miners joining p2pool would of course cry out "I have 1gh, found three shares, the pool got a block but I only got a few cents, scam!!". If people constantly ask why they dont get paid out even though they found 100 shares [diff 1, that is], start explaining them that it takes days, weeks to have a steady income per block.. ;-)
Anyway, sounds good to me, I would do either way, 24h-averages or 7d-averages.
Ente
And, if we all switch to the new version/chain/fork at the same time, we wouldnt receive less per found block, since we all begin at zero.
New miners joining p2pool would of course cry out "I have 1gh, found three shares, the pool got a block but I only got a few cents, scam!!". If people constantly ask why they dont get paid out even though they found 100 shares [diff 1, that is], start explaining them that it takes days, weeks to have a steady income per block.. ;-)
Anyway, sounds good to me, I would do either way, 24h-averages or 7d-averages.
Ente
March 16, 2012, 08:24:23 AM
So, given that we need high difficulty shares, what stops p2pool from using not what a miner produces in a day, but in a week?
This would require some more storage, maybe some more calcs, but should smooth out miner's variance.
Am I wrong as usual?
spiccioli
This would require some more storage, maybe some more calcs, but should smooth out miner's variance.
Am I wrong as usual?
spiccioli
No not wrong but:
a) it will require a hard fork
b) it will mean new users see reduced compensation for 7 days until they have the expected # of shares based on their hashing power.
Maybe 72 hours would be a good compromise for "b" but you still have the issue of "a".
DAT,
if a hard fork is the cost, I'd pay it.
Having less miner variance would encourage people to switch to p2pool, I think.
spiccioli
March 16, 2012, 08:17:53 AM
So, given that we need high difficulty shares, what stops p2pool from using not what a miner produces in a day, but in a week?
This would require some more storage, maybe some more calcs, but should smooth out miner's variance.
Am I wrong as usual?
spiccioli
This would require some more storage, maybe some more calcs, but should smooth out miner's variance.
Am I wrong as usual?
spiccioli
No not wrong but:
a) it will require a hard fork
b) it will mean new users see reduced compensation for 7 days until they have the expected # of shares based on their hashing power.
Maybe 72 hours would be a good compromise for "b" but you still have the issue of "a".
March 16, 2012, 08:13:16 AM
So, I understand variance and all, but is it normal for me to have 3 hour stretches with no p2pool shares found with 2.4 GH/s and expected time to share of 20 minutes?
I have incoming peers, Local hashrate is also about right, local dead on arrival are about 1-2%, efficiency is usually right around 100% or higher.
Is there something off somewhere or is this just variance again?
My node was getting all your luck! I have incoming peers, Local hashrate is also about right, local dead on arrival are about 1-2%, efficiency is usually right around 100% or higher.
Is there something off somewhere or is this just variance again?
I couldn't run my miners for most of yesterday, but had someone mining on it at about 500Mhash/s for most of the day and they found 6 shares in the time it took me to go for lunch
In all seriousness though, if what you describe's happened for a few days in a row, I'd be really concerned. If it's one day, meh, probably really bad variance.
It seems to me that we need some way to mitigate the double variance to which p2pool miners are subject.
We have pool as a whole variance, but also miner variance, for example today my payouts
So, given that we need high difficulty shares, what stops p2pool from using not what a miner produces in a day, but in a week?
This would require some more storage, maybe some more calcs, but should smooth out miner's variance.
Am I wrong as usual?
spiccioli
March 16, 2012, 02:56:12 AM
So, I understand variance and all, but is it normal for me to have 3 hour stretches with no p2pool shares found with 2.4 GH/s and expected time to share of 20 minutes?
I have incoming peers, Local hashrate is also about right, local dead on arrival are about 1-2%, efficiency is usually right around 100% or higher.
Is there something off somewhere or is this just variance again?
My node was getting all your luck! I have incoming peers, Local hashrate is also about right, local dead on arrival are about 1-2%, efficiency is usually right around 100% or higher.
Is there something off somewhere or is this just variance again?
I couldn't run my miners for most of yesterday, but had someone mining on it at about 500Mhash/s for most of the day and they found 6 shares in the time it took me to go for lunch
In all seriousness though, if what you describe's happened for a few days in a row, I'd be really concerned. If it's one day, meh, probably really bad variance.
March 16, 2012, 02:41:13 AM
So, I understand variance and all, but is it normal for me to have 3 hour stretches with no p2pool shares found with 2.4 GH/s and expected time to share of 20 minutes?
I have incoming peers, Local hashrate is also about right, local dead on arrival are about 1-2%, efficiency is usually right around 100% or higher.
Is there something off somewhere or is this just variance again?
I have incoming peers, Local hashrate is also about right, local dead on arrival are about 1-2%, efficiency is usually right around 100% or higher.
Is there something off somewhere or is this just variance again?
March 15, 2012, 01:29:09 PM
Looks like I answered my own question. I checked out /graphs and now the Local graphs show another line item in a different color for each miner! That is so cool!
March 15, 2012, 01:23:58 PM
Looking at the Web Interface documentation here: https://en.bitcoin.it/wiki/P2Pool#Web_interface
I see:
Also, snooping around in graphs.py, I see references to localminer which appears separate from the localrate graphs. Can we get graphs for each individual miner connected to our p2pool node? That would be awesome!
I see "Worker password: xxxxxxxxxxxx (only required for generating graphs)" when p2pool starts up. I used that password as the password for my miners and now I see some rrd files in p2pool/data/bitcoin/. How do I display those in the web interface?
I see:
Quote
/graphs - RRD graphs of past hash rates for pool and local miners. Look for "VIP password" in P2Pool's output when it starts and use that for your miners to get graphs for individual miners.
Also, snooping around in graphs.py, I see references to localminer which appears separate from the localrate graphs. Can we get graphs for each individual miner connected to our p2pool node? That would be awesome!
I see "Worker password: xxxxxxxxxxxx (only required for generating graphs)" when p2pool starts up. I used that password as the password for my miners and now I see some rrd files in p2pool/data/bitcoin/. How do I display those in the web interface?
March 15, 2012, 11:05:37 AM
Love you!
Everyone please mine with me and support P2Pool! https://bitcointalksearch.org/topic/closed-8-ghs-bit-pop-p2pool-zero-fee-auto-payments-oregon-usa-68743
Everyone please mine with me and support P2Pool! https://bitcointalksearch.org/topic/closed-8-ghs-bit-pop-p2pool-zero-fee-auto-payments-oregon-usa-68743
March 15, 2012, 10:51:13 AM
Last q, even if I explicitly set payout add, i dont have to do a thing to let users successfully use their add as a username?
Nope, username always overrides the set payout address unless there's a fee set. March 15, 2012, 10:42:39 AM
Last q, even if I explicitly set payout add, i dont have to do a thing to let users successfully use their add as a username?
March 15, 2012, 10:40:08 AM
Can anyone attack me by making my p2pool public? Or any private info leaking through the web interface?
http://bitpoppool.geekgalaxy.com:9332/fee
No, nothing there needs to be private (it does expose your payout address though). The only possible attack is a DoS.http://bitpoppool.geekgalaxy.com:9332/fee
No, you need to run "bitcoind sendmany XXX" to make the transaction. That just creates the output list.
March 15, 2012, 10:38:54 AM
Is this actually sending?
http://bitpoppool.geekgalaxy.com:9332/patron_sendmany/.001
http://bitpoppool.geekgalaxy.com:9332/patron_sendmany/.001
March 15, 2012, 10:36:53 AM
Can anyone attack me by making my p2pool public? Or any private info leaking through the web interface?
http://bitpoppool.geekgalaxy.com:9332/fee
http://bitpoppool.geekgalaxy.com:9332/fee
March 15, 2012, 10:35:32 AM
Thanks!
Where is a list of url resources I can look at such as http://bitpoppool.geekgalaxy.com:9332/fee
https://en.bitcoin.it/wiki/P2Pool#Web_interface March 15, 2012, 09:55:20 AM
Where is a list of url resources I can look at such as http://bitpoppool.geekgalaxy.com:9332/fee
https://en.bitcoin.it/wiki/P2Pool#Web_interface 
March 15, 2012, 09:34:31 AM
I've used pypy for a couple of days, it works ok, I don't see a big difference in CPU usage compared to python, but it has a problem: it is not compatible with pygame so you cannot show/generate graphs.
Bummer about both. I don't use the graphs so hadn't noticed that pygame wasn't compatible. But if it doesn't offer any significant CPU efficiency gain, it's probably irrelevant.
March 15, 2012, 08:37:31 AM
Where is a list of url resources I can look at such as http://bitpoppool.geekgalaxy.com:9332/fee
March 15, 2012, 08:06:45 AM
No you shouldn't have to port anything. PyPy should just work with normal Python code. It's simply a different implementation of the standard Python interpreter. No difference in the language it interprets, just how it does it.
For one thing it has a built-in Just-in-time compiler powered by LLVM, which means that instead of the writers of a Python implementation needing to worry about how to create an efficient Just-in-time compiler, the work of people from the LLVM project, who are already experts in constructing such a compiler, can be leveraged.
For one thing it has a built-in Just-in-time compiler powered by LLVM, which means that instead of the writers of a Python implementation needing to worry about how to create an efficient Just-in-time compiler, the work of people from the LLVM project, who are already experts in constructing such a compiler, can be leveraged.
This made me curious, so I am running p2pool with pypy as we speak. It seems to be working fine, but I'll watch it for a day or two and see if there are any abnormalities.
twmz,
I've used pypy for a couple of days, it works ok, I don't see a big difference in CPU usage compared to python, but it has a problem: it is not compatible with pygame so you cannot show/generate graphs.
spiccioli
Jump to: