Bitcoin Forum

Bitcoin => Development & Technical Discussion => Topic started by: Synchronice on July 15, 2023, 11:25:50 AM



Title: Create a vanitygen address via Cloud GPUs - recommendation?
Post by: Synchronice on July 15, 2023, 11:25:50 AM
I want to create vanity bitcoin addresses via cloud GPUs. Just wanna play with the power of cloud GPUs and their capability to produce long customized bitcoin addresses, so,
are there any cloud companies that let you to use a lot of powerful GPU servers and pay according to the load?
At the moment, some companies that I have seen, are:
Vast.ai
Cherry Servers
RunPod
Lambda Labs
AWS and Google Cloud are excluded.

Which one do you recommend from the list? Or outside of the list?

In the best case, I'm looking for something similar to nicehash where I'll be able to rent hash power and direct it the way I want, in my case, on generating of vanity addresses.

P.S. I am not looking for monthly rents, I just wanna rent for some hours or days in worst case.
P.P.S. I know it's not a safe option to generate Bitcoin addresses outside of airgapped computer, I'm just having some fun, that's all.


Title: Re: Create a vanitygen address via Cloud GPUs - recommendation?
Post by: Cricktor on July 15, 2023, 01:06:40 PM
With the split-key technique you can still create safe vanity addresses and doing the address mining on some remote cloud computers wouldn't compromise the safety of your vanity address(es). Of course, the recombination of the split keys needs a secure environment controlled by yourself alone.

Sorry, I can't contribute though to your questions where you can use GPU power on demand and in the way necessary to perform address mining.


Title: Re: Create a vanitygen address via Cloud GPUs - recommendation?
Post by: ranochigo on July 15, 2023, 01:37:39 PM
A lot of the Cloud GPU are machine learning workload optimized, think large and fast amount of RAM to store them in the memory for faster access. You would probably want to try Vast.AI, had some experience with it previously for an unrelated ML project and it worked like it should, though it is definitely a lot more expensive.

AWS EC2 would be a good idea as well though.