Pages:
Author

Topic: Is there a way to group all my machines into one ? (Read 347 times)

member
Activity: 434
Merit: 52

Nope.  Solves the problem of hanging rigs when you are not on site.  It's pretty obvious what he needs to do is stop building 2 gpu rigs.

I use CPU hardware.  Servers, some small some big.  I build what I can afford, and use what I have.  My small machines are used for testing, the big ones are for production after I verify my tests on a small scale. 

"build bigger rigs" is an easy answer.  Buying new / different hardware is not an option.  I did not ask "how do I get more hash power ?".  Please remain focused on the topic.


Ah. CPU mining. Well, yes, you CAN build a cluster and control them that way, but I don't think it's as productive. I looked into it when I bought my little mini-army of Brix i5s and i7s and my take away was that it is no more efficient, and most people said it was slightly less. Might be worth it for the convenience, but writing automation scripts is just as easy AND allows you to control each individually if they go down or need repair. YMMV, and of course half the fun is messing with the machines, so do post your results!
newbie
Activity: 8
Merit: 0

[/quote]

In that case.  It seems not worth the time or hassle of maintaining them.  Good luck.
[/quote]

I am a software developer, it's what I do.  Dumping money into rigs is my last step, I am not in a hurry to make a mistake.  Writing code and analyzing tests is not a hassle for me, it's my pleasure and experience.
newbie
Activity: 8
Merit: 0


look for a stratum proxy for your algo, there's one for cryptonight here:

https://bitcointalksearch.org/topic/stratum-proxy-for-cryptonight-2654871


[/quote]

Finally - yes, this was the type of answer I am looking for.  Thank you.

In case anyone else was actually interested, I found:

"XMR-NODE-PROXY"

Quote
allows you to merge multiple Monero mining rigs into one, which helps lower the load on pools, reduces bandwidth used and allows you to use a lower difficulty setting. When using a proxy, the mining pool that you connect to see only one miner.

The key word here was "LOWER DIFFICULTY SETTING".   By doing so, allows the smaller machines to become more efficient.  
member
Activity: 644
Merit: 24

Nope.  Solves the problem of hanging rigs when you are not on site.  It's pretty obvious what he needs to do is stop building 2 gpu rigs.

I use CPU hardware.  Servers, some small some big.  I build what I can afford, and use what I have.  My small machines are used for testing, the big ones are for production after I verify my tests on a small scale. 

"build bigger rigs" is an easy answer.  Buying new / different hardware is not an option.  I did not ask "how do I get more hash power ?".  Please remain focused on the topic.





In that case.  It seems not worth the time or hassle of maintaining them.  Good luck.
newbie
Activity: 8
Merit: 0

Nope.  Solves the problem of hanging rigs when you are not on site.  It's pretty obvious what he needs to do is stop building 2 gpu rigs.

I use CPU hardware.  Servers, some small some big.  I build what I can afford, and use what I have.  My small machines are used for testing, the big ones are for production after I verify my tests on a small scale. 

"build bigger rigs" is an easy answer.  Buying new / different hardware is not an option.  I did not ask "how do I get more hash power ?".  Please remain focused on the topic.



member
Activity: 644
Merit: 24
so many opinions guys , and no one get it what he means , lol

im sure to know  after 10 seconds what he has in mind

simple as that , did you heard of HPC

High Performance Computing  , for Noobs  simply a Cluster-Enviroment

1 Master , 3 Nodes (slaves) in his case

OS , Application on Master working as server for the Nodes

to go deeper in detail is a bit to much here

if you interested on things such this Search by Uncle Google for "PelicanHPC"



Yeah I get it, but I don't see the need in this application.  He has a very small farm.  Build the ship to fit the crew.  For as small of a crew as you have, they should all fit in one ship.
full member
Activity: 144
Merit: 100
Eager to learn
so many opinions guys , and no one get it what he means , lol

im sure to know  after 10 seconds what he has in mind

simple as that , did you heard of HPC

High Performance Computing  , for Noobs  simply a Cluster-Enviroment

1 Master , 3 Nodes (slaves) in his case

OS , Application on Master working as server for the Nodes

to go deeper in detail is a bit to much here

if you interested on things such this Search by Uncle Google for "PelicanHPC"

member
Activity: 134
Merit: 10
I have 4 machines connecting and each of them produce at a different rate on the same algorithm and pool.  Is there an application I can run on to be a master, and to combine the hash power off to my slave machines?

I would like to just merge the power of all the machines to one.  Have the master application distribute the load internally to my machines and channel it through to the pool.  To the external pool, I would have one machine - but instead I have 4 of them working together.

look for a stratum proxy for your algo, there's one for cryptonight here:

https://bitcointalksearch.org/topic/stratum-proxy-for-cryptonight-2654871

member
Activity: 644
Merit: 24

Never heard of this before.  Build bigger rigs.  Don't have small rigs.  I honestly don't think it makes any difference.  Difference is in the hardware even if they are the exact same model.  Did you think of the other variables that you haven't thought about yet?


Main difference is if one goes down, you still have others mining while you fix it. But yeah, if you trust your gear, you're only saving a little bit of electricity (for my three rigs, combining would probably save 80W) by combining.

Switched PDU solves that problem. 

How so? It removes the need to power the two extra cpus?

Nope.  Solves the problem of hanging rigs when you are not on site.  It's pretty obvious what he needs to do is stop building 2 gpu rigs.
member
Activity: 434
Merit: 52

Never heard of this before.  Build bigger rigs.  Don't have small rigs.  I honestly don't think it makes any difference.  Difference is in the hardware even if they are the exact same model.  Did you think of the other variables that you haven't thought about yet?


Main difference is if one goes down, you still have others mining while you fix it. But yeah, if you trust your gear, you're only saving a little bit of electricity (for my three rigs, combining would probably save 80W) by combining.

Switched PDU solves that problem. 

How so? It removes the need to power the two extra cpus?
member
Activity: 644
Merit: 24

Never heard of this before.  Build bigger rigs.  Don't have small rigs.  I honestly don't think it makes any difference.  Difference is in the hardware even if they are the exact same model.  Did you think of the other variables that you haven't thought about yet?


Main difference is if one goes down, you still have others mining while you fix it. But yeah, if you trust your gear, you're only saving a little bit of electricity (for my three rigs, combining would probably save 80W) by combining.

Switched PDU solves that problem. 
member
Activity: 434
Merit: 52

Never heard of this before.  Build bigger rigs.  Don't have small rigs.  I honestly don't think it makes any difference.  Difference is in the hardware even if they are the exact same model.  Did you think of the other variables that you haven't thought about yet?


Main difference is if one goes down, you still have others mining while you fix it. But yeah, if you trust your gear, you're only saving a little bit of electricity (for my three rigs, combining would probably save 80W) by combining.
member
Activity: 644
Merit: 24
When I compare the ratio of worked performed and accepted shares on my smaller machines compared to my larger ones - the pool seems to sometimes drown out the smaller machines.  A small starving machine with barely any data to process suffers doing nothing but wasting electric.  My idea would be to combine my work into one large loaded stream, to make efficient use of all my machines and not just the largest and fastest.  There are other advantages as well.  





Never heard of this before.  Build bigger rigs.  Don't have small rigs.  I honestly don't think it makes any difference.  Difference is in the hardware even if they are the exact same model.  Did you think of the other variables that you haven't thought about yet?

newbie
Activity: 8
Merit: 0
When I compare the ratio of worked performed and accepted shares on my smaller machines compared to my larger ones - the pool seems to sometimes drown out the smaller machines.  A small starving machine with barely any data to process suffers doing nothing but wasting electric.  My idea would be to combine my work into one large loaded stream, to make efficient use of all my machines and not just the largest and fastest.  There are other advantages as well.  



member
Activity: 644
Merit: 24
I understand what he wants, I just don't know the purpose or benefit that he thinks he will get from it.
sr. member
Activity: 1008
Merit: 297
Grow with community
I've reached 5x times back and forth reading to comprehend what the OP wants. Cheesy

and then suddenly I realized this statement '

Quote
To the external pool, I would have one machine - but instead I have 4 of them working together.

A very simple solution to your request my friend, Just give "1" Similar worker name to each of your Rig, you will now see a combine hash power to the pool.  Grin
member
Activity: 644
Merit: 24
I'm trying to figure out why you even want to do that?
legendary
Activity: 1498
Merit: 1030
There is no reason to do what you want to do, even if you COULD do so.
hero member
Activity: 756
Merit: 560
I am not sure I am understanding your question. If you use the same worker name on all of the machines they will show up on the pool as a single user.

There is no way to merge anything, each individual card has its own mining thread, even in a single machine there is nothing 'shared'
Pages:
Jump to: