4

Many bloggers have pointed out that it is almost impossible to sync an ethereum archive node using a HDD.But most of those blogs were published in 2018.

I've seen several version updates of geth since then, and some improvements had been made to sync an archive nodes.

So I would like to ask if anyone has successfully synched an ethereum full archive node with HDD recently?Use either geth or parity.

蒋心健
  • 71
  • 3

2 Answers2

3

No, you need to use an SSD.

Q: Wait, so I can’t run a full node on an HDD?

A: Unfortunately not. Doing a fast sync on an HDD will take more time than you’re willing to wait with the current data schema. Even if you do wait it out, an HDD will not be able to keep up with the read/write requirements of transaction processing on mainnet.

You however should be able to run a light client on an HDD with minimal impact on system resources. If you wish to run a full node however, an SSD is your only option.

From the Geth FAQ expanded in related answer about fast sync.

eth
  • 85,679
  • 53
  • 285
  • 406
1

By using a server having very large amount of RAM, it is possible. We got a clue from https://blog.ethereum.org/2019/07/10/geth-v1-9-0/ :

The reason for the higher disk IO is due to using less memory for caching, having to push more aggressively to disk.

That means with more RAM, disk IO can be reduced. We were able to sync archive node on 10K SAS HDD with approx. 700+GB RAM and --cache 614400 --snapshot=false, though it took some months to get fully synced :)

curious
  • 21
  • 2
  • Is the economically better or worse option? Would 4 TB SSD be less expensive than 1TB RAM? – Mikko Ohtamaa Aug 04 '22 at 12:04
  • Also you might get much better results with Erigon than geth: https://github.com/ledgerwatch/erigon – Mikko Ohtamaa Aug 04 '22 at 12:08
  • Sooo, Ethereum bad, Bitcoin good? Just kidding (/me ducks) – Justin Goldberg Aug 04 '22 at 13:54
  • 1
    @Mikko-Ohtamaa That is actually an interesting debate. 1TB RAM would be a one time purchase. Given the current ~13TB size of of archive node (and size will always keep increasing), more than 13TB (nvme) SSD will be required. Once the node is synced, not that much resources are required and excess RAM may be allocated to some other server :) – curious Aug 05 '22 at 14:12