WARN near_store::db::rocksdb: target="store::db::rocksdb" making a write batch took a very long time, make smaller transactions!

[MaxZubrytskyi]

Describe the bug
I'm starting fresh node, v2.0.0, config from off url, downloaded latest snapshot, my machine is pretty good one, 64 cores, 256GB RAM and few NVME in RAID, first it does download 1.4TB snapshot, then start pulling latest blocks to sync but warning started and it is keep on "downloading latest blocks" for 12h alredy (1.4TB snapshot was pulled within 4h)
warning message like

WARN near_store::db::rocksdb: target="store::db::rocksdb" making a write batch took a very long time, make smaller transactions! elapsed=178.067339ms backtrace=   0: <near_store::db::rocksdb::RocksDB as near_store::db::Database>::write
   1: near_store::StoreUpdate::commit
   2: near_chain::store::ChainStoreUpdate::commit
   3: near_chain::chain::Chain::postprocess_block
   4: near_client::client_actor::ClientActorInner::try_process_unfinished_blocks
   5: near_client::client_actor::ClientActorInner::check_triggers
   6: <actix::address::envelope::SyncEnvelopeProxy<M> as actix::address::envelope::EnvelopeProxy<A>>::handle
   7: <actix::contextimpl::ContextFut<A,C> as core::future::future::Future>::poll
   8: tokio::runtime::task::raw::poll
   9: tokio::task::local::LocalSet::tick
  10: tokio::task::local::LocalSet::run_until::{{closure}}
  11: std::sys_common::backtrace::__rust_begin_short_backtrace
  12: core::ops::function::FnOnce::call_once{{vtable.shim}}
  13: std::sys::pal::unix::thread::Thread::new::thread_start
  14: start_thread
  15: clone

To Reproduce
just start node

Expected behavior
fully synced and start work

Version (please complete the following information):

• neard (release 2.0.0) (build unknown) (rustc 1.78.0) (protocol 69) (db 39)
  features: [default, json_rpc, rosetta_rpc]
• mainnet

[MaxZubrytskyi]

update: it does finished sync to latest block within 20h and once sync it does randomly throw this warning and right after it it keeps downloading last 10-30 blocks, issue is intermittent and occurs every 10-15m

also noticed huge disk IO usage.
maybe there are some settings which could help to reduce disk load?

[MaxZubrytskyi]

cpu: AMD EPYC 7502P 32-Core Processor
RAM: 256GB (8x32GB) Speed: 3200 MT/s Manufacturer: Micron Technology
NVME: Micron_7450_MTFDKCC7T6TFR

[MaxZubrytskyi]

we did replace server with new one, which do have gen V4 NVME slots (which does increase IOPS for us, as we does ahve gen3 slots with gen4 nvme, so that nvme were unable to use max performance ) and issue is gone, so for all who faced same: it was a disk pressure issue.