Topic: Bitcoin blockchain sync very slow on HDD (Read 573 times)

You need to reindex, which takes 3 to 5 hours and rebuilds chainstate

A clean shutdown in the middle of IBD will write chainstate to disk. Then if you backup chainstate plus the most recent BLKnnn.DAT, you have a checkpoint where chainstate is synced as at the most recent block in that BLK file. This assumes none of the older now-closed BLK files ever gets corrupted. Chainstate is about 12GB and BLK files are never bigger than 128MB, so it's a fairly light backup

Some notes:

The lopp.net blog has some articles on sync speed

If you have 16GB RAM do not allocate all of it to dbcache. I allow 1.2GB for OS and other functions, and I don't use GUI. This means set dbcache to 14800

I've only ever used HDD
With 4GB RAM and an I3 CPU, I set dbcache to 2800 and the IBD took 12 days
With 8GB RAM, I set dbcache to 6800 and the IBD finished in 55 hours
With 32GB RAM, I set dbcache to 16000 and the IDB finished in 16 hours
I have 2 HDD. It is faster to have chainstate on one HDD and blocks on a different HDD
According to the documentation, dbcache can't be more than 16000
Speed is best monitored as transactions per minute/hour, not blocks
I never use the GUI
Blocks are only written sequentially, never updated after closing each BLK**.DAT
Chainstate is constantly inserted and deleted. Using bigger dbcache (up to 16GB) stores more of chainstate in RAM, reducing the time-consuming insert/delete writes to disk
IBD does not verify txinput signatures if "assumevalid" is enabled, until the most recent blocks. IBD slows down a lot when it gets to this block, but by then it's nearly finished

Things not yet tried ...
Chainstate is about 12GB, so it would be cheap to use a small SSD for chainstate and keep blocks on HDD

...

Actual numbers are useful, but few people bother to make measurements. This means the generic "get a SSD" advice is based in ignorance
The biggest determinant is dbcache. One of the recent lopp.net articles has some measurements
Most people asking about slow node initialization do not specify how long it's taking, or what they expected, do not mention how much RAM they have, and never mention dbcache
Most "how to" articles fail to mention the initialization time. This is because most of those bloggers never actually finish the IBD, and do not admit it

I don't think Bitcoin Core do other intensive task (which use lots of RAM) during sync. But other than that, your suggestion makes sense, although it's on safer side.

The reason for such value is very simple. Bitcoin Core runs on machine with OS which needs RAM for its routine operations. Besides node itself perform plenty of tasks outside those ones related with its databases. Sure this value of 1/4 RAM is machines-average one, thus if one needs the best value for the given device/OS he must analyze the RAM usage before starting the synch procedure.  But if he prefer not to bother himself with such kind of analysis 1/4 RAM would be the best option for him.

I've seen some people make this suggestion, but what are the reason behind choosing 1/4 RAM? I would rather say majority available RAM capacity or current serialized UTXO size[1].

[1] https://statoshi.info/d/000000009/unspent-transaction-output-set

There are: dumptxoutset and loadtxoutset RPC commands, maybe these are the commands that you saw?
But dumptxoutset doesn't actually dump to the original chainstate, it just creates a file that can be loaded with loadtxoutset.

Although, it's useful for anyone who want to use the wallet without waiting for IBD since the loaded UTXO set (from someone/backup) will be used temporarily instead.
(currently only in RegTest, TestNet or Signet; ref: https://github.com/bitcoin/bitcoin/blob/master/src/kernel/chainparams.cpp#L176-L178)
Ref: https://github.com/bitcoin/bitcoin/blob/master/doc/design/assumeutxo.md

True. I myself have experienced   the significant  reduction of synch time after  replacing my HDD (7200 rpm) by NVMe SSD.

 I/O value does matter when the case in hand is blockchain synchronization. Those who deny this fact were never fiddling with SSD.

BTW, the optimal value of dbcache is 1/4 RAM.

Let's say you do this, and restart Bitcoin Core. Then, your system crashes and chainstate is messed up. Don't you still need to rebuild it from scratch? Or can it continue from the block where it was last saved to disk?

It doesn't. Chainstate and mempool are different. I'm not aware of any commands that flushes the chainstate automatically because the frequency should presumably should be governed by the dbcache capacity.
So long as you don't suffer any hiccups during the IBD. It would be quite a pain to rebuild the chainstate again.

I think there is a command in Bitcoin Core's RPC that will allow you to manually flush things like the chainstate to the disk though? savemempool or something like that.  It obviously saves the mempool data but I don't know if it also saves the chainstate as well.

You could do something like gracefully shutting down the node after the IBD and then restarting it though.

While that is mostly true, I still find I/O to be the bottleneck at times after the default assumevalid block. With the default dbcache, my synchronization time is around ~9 hours but 20GB of dbcache brings it down to 7.5hours. That is with NVMe, when comparing SSD and HDD, the difference is far bigger.

I've also discovered one of the cons of maintaining the entire chainstate (maxing out the dbcache) recently. Having all of the chainstate in the dbcache will result in the chainstate having to be rebuilt upon ungraceful shutdown due to it not being flushed to disk. It also depends if it really is feasible to add more ram for IBD, if they have no use for all that extra ram.

With sufficient RAM, the type of disk barely matters. All it does is save the blocks, and save chainstate when you shut it down.

My experience is that bandwidth isn't the key factor with your synchronization, because your peers are probably not going to be able to saturate your bandwidth. I've had similar performance with 2GBPS, 1GBPS and 200mbps. Most significant improvement probably came around when I got an NVMe.

That's a pretty good system.  I like the size of it, too.  

My latest build was a on 1u server, which can be cumbersome if you're not set up to house servers.  Enterprise servers that have met their End of Life cycle get replaced regularly, and can be had for pretty darn cheap on Ebay.  With a 256 megabaud internet connection I was able to sync the blockchain in under 6 hours.  Syncing Monero still took a few days for some reason.

System: HP ProLiant DL360 G7 Server
Processors: 2x Intel Xeon X5650 @ 2.67GHz
Ram: 18x 4GB DDR3 (74GB total)
Storage: 4x 1TB SSD Raid 5 (2.8TB usable storage)
OS: Debian 12

The total build cost me about $330 a little more than a year ago.

Yeah I am thinking of doing it using this ssd

https://www.samsung.com/us/computing/memory-storage/solid-state-drives/980-pro-w-heatsink-pcie-4-0-nvme-ssd-2tb-mz-v8p2t0cw/?

Oh BTW I found a pretty good deal on an elite mini 800 g9 with an i7 14700T cpu 64gb ram and a 4tb nvme.2 ssd the eth port has a 1Gbps line

https://www.ebay.com/itm/126560449790?

this pc would pretty much be a perfect node for the next 5 years

A 1Gbps line that consistently hits that speed can finish the initial block download in a couple hours as long as you also have an enterprise HDD (or SSD).

May I ask why? I'd use a VM if I don't trust the software (such as Fork wallets). For daily wallets it's much more convenient to run them on your main system (or as a different user to separate file access).

Got SSDs and tested. My conclusion is here.


So,

With HDD and maybe 4GB for dbcache, 2 to 3 weeks in 12 hours. GUI stalled badly.

With HDD and 16GB for dbcache, 5 weeks in 2 to 3 hours. GUI still stalled badly I think.

With SSD and 16GB for dbcache, 2 weeks in less than one hour. Maybe less than 30 minutes? And GUI felt less unresponsive.


Roughly, 10 times speed increase with bigger dbcache, another 2x increase with SSD and same dbcache.

My benchmarks aren't equal because I didn't have much blockchain left when I got RAM and SSD upgrades. But roughly.


With Monero,

RAM upgrade didn't help.

With HDD speed was 500 to 600 blocks/h during the final 50000 to 100000 blocks.

With SSD, speed was around 5000 blocks/h, a bit more or a bit less at times. So 10x increase in speed.


A year ago perhaps, when I benchmarked file systems, I got over 5000 blocks/h on HDD. Up to 9000 blocks/h at times. But Monero has some kind of checkpointing that speeds up syncing I think? So these cannot be compared to today's results?


It must be said that this computer is not new, and my cryptocurrency stuff are running in a virtual machine. The disk is passed to the virtual machine and I don't know how much overhead this brings.

well i don't know what world you are living in but a core i5 is not excessive and neither is a 2TB storage space especially if 25% of that is being hogged up by the blockchain...

This is still excessive for syncing a Bitcoin blockchain. All it does is shave a few hours off the total sync time. I'd base my computing requirements on my daily needs, not on a one-time event.

that pc was the best i ever owned.

it was serious money the threadripper alone was 2500


you don’t need all that.

Any i5 intel  8500 or newer

16 gb better to have 32gb

and any good  2tb ssd. nmve 2

oh. 200. mh internet .

I run full node, so i could run gettxoutsetinfo "none" block_height to get such data. There are few website which monitor total UTXO such as,
https://statoshi.info/d/000000009/unspent-transaction-output-set
https://www.blockchain.com/explorer/charts/utxo-count
https://utxo-stats.com/

If you read discussion about IBD with Raspberry Pi or old device, you might notice people report about slow IBD due to low RAM.

That point must be after the specific block hash set here: github.com/bitcoin/bitcoin/blob/c4d45b695ef27e21d35e886b08887e2ecf272ce7/src/kernel/chainparams.cpp#L108
But it should've recovered after it's finished with all the heavy signature script verifications.

Well, a "pro" will ask for the debug.log to look for the root of the issue instead of suggesting solutions that may not be necessary.
The reason why you're not asked for logs here is because slow sync is pretty much hardware and/or database cache issue
and the "dbcache" solution that you found is consistent in topics like this.

If you followed this format (sticky topic): [READ BEFORE POSTING] Tech Support Help Request Format
Someone pro can easily point that your ram and dbcache settings can be improved just from looking at the logs and the provided info.

It's natural that with a same cake, if there are more people know about the cake, want to have pieces of the cake, each will have smaller piece. People who have bitcoins need to take care of their bitcoins, hold their coins tightly and sit down to think of solid plans for investment, holding and personal, family financial plans.

They need very good plans to avoid of any situation that forces them to sell their bitcoins and turn their pieces to be smaller and smaller ones with time. They had chances to join the market earlier than late comers but if they don't have proper plans, they will end with no differences to late comers.

Bitcoin: Addresses Holding > X BTC by Year

I don't keep track of UTXOs, but I have data on funded addresses. Apart from small temporary reductions (usually due to consolidation at low fees), they're generally growing in number.
The way I see it: Bitcoin gets more expensive, and more people own smaller and smaller bits of it.

It depends on the transactions. If the future transactions create more UTXO than spend, then it will be worse over time, yes.

I believe it's just your HDD. It's extremely slow to download and index all that information on an HDD. Have you tried SSD?

How do you know that the UTXO keeps growing? Are there stats that support this?

I just got mine sync and its 660GB .... done in 5 days with a 3MBPS line on my phone data.
Just 16gb amd ryzen 9 7950 and 1tb ssd ext drive.

Yea, i did set the cache to 1800MB compare to originally 450MB. It does help. Even on my raspberry pi 5, i did sync it within 5 days on the same ext drive.

same core version 27.0. I think it could be the cache 450MB default setting. Try changing it and trying different setting. Hope it helps.

Based on my experience, the GUI would be responsive again and show latest information. Although if you need to know latest sync progress, you could open debug.log and look for newest line which contain text "UpdateTip".


There are other parameter you could try, such as --block-sync-size, --fast-block-sync and --db-sync-mode. If you have question specifically about Monero, you should create new thread/reply on different board.

Last time I synced Monero (on a server with 32 GB RAM, and HDD), the first thing it told me was to get a SSD. I couldn't do that, and eventually removed Monero because even after it was synced, the entire server felt "sluggish" while it was running.

how much did that setup cost you? i'm guessing around $2500. most people would be better off just buying bitcoin instead of trying to spend that money to download the blockchain. it's just going to be an ever increasing game of having to throw more and more hardware at the problem big waste of money unless you just have that stuff lying around from something else...

Thank you, for answers.

Finally, actually, I got "pro level" answers instead of just "works fine for me", "can u show ur logfiles", "just get SSD bruh". 

I got a cheap RAM upgrade, set dbcache to 16GB (was 4GB or 6GB), started syncing.

At some point the GUI had stalled very badly, I could do nothing so I shut down Bitcoin-qt and it shut down cleanly. When I restarted it the blockchain was in fully synced state. So my remaining 5 weeks were synced in 2 to 3 hours, much faster than before.

I don't know if the GUI would have recovered if I had waited for some more time. Maybe I tried to use it just at the critical moment. It's unknown for now.


Monero did not get faster than before and the last 50000 to 100000 blocks seem to be much slower than anything before. I didn't find any option similar to dbcache in Bitcoin. 600 blocks/h is the speed now. Used to be 2200 blocks/h for quite long. I benchmarked different file systems and certain types were occasionally faster than that but fell to the same slow level at some point.

Maybe it's SSD time next.

Short answer, yes. Not only each block become bigger, total UTXO keep growing.

Edit: I assume "gets longer" refers to newer blocks.


AFAIK no. New stuff (such as SegWit) solve few problem including quadratic verification time.


No. Jameson Lopp did benchmark which prove the opposite on https://blog.lopp.net/bitcoin-core-performance-evolution/.


That's true, it's known issue for few years. See this issue, https://github.com/bitcoin-core/gui/issues/299.

a decent i5 gen cpu say i5 8500 or newer
a 1 tb samsung ssd or bigger
16gb ram or more
a decent internet connection say 100Mbps or more


if you have all of the above you will be under 1 day


obvious that

an i7 14700
a 2tb nvme2 quality ssd
64gb ram
and a 1Gbps internet

will be faster.

In fact the second setup may be faster than the offered download speed.

I have used 200Mbps and
128gb ram
1tb nvme2 ssd
threadripper cpu

My setup was able to do 190Mbps constantly which makes me suspect my internet connection was my bottle neck

I will soon get 1gb internet and try to see how fast I can do this.

You have 2 bottlenecks: RAM, and HDD speed. Both are more or less interchangeable: with 32 GB RAM and 12000 MB dbcache, your sync will go just fine. With 8 GB RAM and an SSD, the SSD is still the bottleneck. I tested both of these examples myself.

Ten years ago, the entire blockchain was only 20 GB.

Yes. The chainstate directory keeps growing, and requires more and more RAM (or a LOT of disk activity) to verify the download.

Nope. Older versions were much slower to sync.

Monero is even more demanding than Bitcoin Core, despite a much smaller blockchain.

TL;DR: get an SSD.
If you have a small SSD, put the chainstate directory on it. That will make a huge difference for Bitcoin Core. If you don't have an SSD yet, buy one. Get a decent one for best performance. It's going to largely improve your entire computing experience.

It should, but the chain is not increasing as such a rate that will cause an significant lag in download time. Could be from the internet connection.

Nothing happened on the protocol level to make it harder either.

- Jay -

In the last 10 years there're fewer transactions in a block... now there's  more.
Use the current version check the latest news section  of the forum for the torrent file,
Then if you are using bitcoin-qt GUI  check the dbcache increase it, it could make it sync process more  faster(note the value is in MB ) this should  be done carefully  in accordance  to your  ram  capacity .
Not really but yes and might depends on the hardware , people can still get it done in just 2~4 days.
Sometimes when when block index is writing to disk you might experience  that and you would think it's just stall

I'm syncing my Bitcoin wallet blockchain using latest Bitcoin Core on Linux.

It progresses very very slowly, right now 2 to 3 weeks in 12 hours.

CPU and net usage stay low most of the time, problem is not there (most likely). HDD usage is constant, I think that's probably the bottleneck. It's a 7200rpm standard desktop HDD.

10 years ago I could sync the blockchain on a slow 5400rpm laptop HDD without major issues.

Time to buy big SSD perhaps.

But the interesting question is,
Does syncing get harder over time as the blockchain gets longer?

Or has something happened to the protocol so it would be harder now?

Or has the code quality of Bitcoin Core gone down?


The GUI also stalls badly during sync. I don't remember when this problem began but it did not always exist.

Syncing monero is also extremely slow and it has been so as long as I can remember.  But that's another story.