2

I've reviewed the response here: How do pool operators do their hashing? and acknowledge that the ratio of hashing required by pool operators is incomparable to the miners.

I still wonder what hardware do Mining pools employ to perform the calculation of the hashmerkleroot for every miner? Would there be any performance improvement in utilising specialised hardware vs a generic yet powerful CPU? (especially when there are an average of 2000 transactions per block each being around 600 bytes in size)

1 Answer 1

3

They don't have any special hardware, as pools don't meaningfully produce merkle roots in any volume. Since the original answer in 2013 which pertained to GetWork, mining has moved entirely to using the Stratum protocol which has the mining client generate their own merkle roots for proof of work, based on a template that is provided by the pool and is the same for all clients. Modern miners can be exhausting more than 10,000+ nonce ranges a second so the previous system is simply infeasible to use.

5
  • Thanks, so by mining clients you mean the miner ASICs perform the hashmerkleroot computation? Sep 7, 2021 at 11:01
  • Whatever local hardware is controlling the ASICs, usually a normal machine running linux, sometimes offloading some of the hashing to a FPGA.
    – Claris
    Sep 7, 2021 at 11:02
  • Thanks. Makes some sense but I've seen cases where the Antminer S19 pro is practically plug and play to a mining pool without anything else controlling/overseeing it? Surely this means that the hashmerkleroot is being computed by by the Antminer S19 pro ASIC? Sep 7, 2021 at 20:43
  • Antminer machines contain an entire linux computer, the ASIC is not producing the merkle root.
    – Claris
    Sep 8, 2021 at 0:08
  • Thanks. All makes sense now. appreciate it! Sep 8, 2021 at 9:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.