Re: backup manifests

Hi,

On 2020-03-26 14:02:29 -0400, Robert Haas wrote:
> On Thu, Mar 26, 2020 at 12:34 PM Stephen Frost <sfrost(at)snowman(dot)net> wrote:
> > Why was crc32c
> > picked for that purpose?
>
> Because it was discovered that 64-bit CRC was too slow, per commit
> 21fda22ec46deb7734f793ef4d7fa6c226b4c78e.

Well, a 32bit crc, not crc32c. IIRC it was the ethernet polynomial (+
bug). We switched to crc32c at some point because there are hardware
implementations:

commit 5028f22f6eb0579890689655285a4778b4ffc460
Author: Heikki Linnakangas <heikki(dot)linnakangas(at)iki(dot)fi>
Date: 2014-11-04 11:35:15 +0200

Switch to CRC-32C in WAL and other places.

> Like, suppose we change the default from CRC-32C to SHA-something. On
> the upside, the error detection rate will increase from 99.9999999+%
> to something much closer to 100%.

FWIW, I don't buy the relevancy of 99.9999999+% at all. That's assuming
a single bit error (at relevant lengths, before that it's single burst
errors of a greater length), which isn't that relevant for our purposes.

That's not to say that I don't think a CRC check can provide value. It
does provide a high likelihood of detecting enough errors, including
coding errors in how data is restored (not unimportant), that you're
likely not find out aobut a problem soon.

> On the downside,
> backups will get as much as 40-50% slower for some users. I hope we
> can agree that both detecting errors and taking backups quickly are
> important. However, it is hard for me to imagine that the typical user
> would want to pay even a 5-10% performance penalty when taking a
> backup in order to improve an error detection feature which they may
> not even use and which already has less than a one-in-a-billion chance
> of going wrong.

FWIW, that seems far too large a slowdown to default to for me. Most
people aren't going to be able to figure out that it's the checksum
parameter that causes this slowdown, there just going to feel the pain
of the backup being much slower than their hardware.

A few hundred megabytes of streaming reads/writes really doesn't take a
beefy server these days. Medium sized VMs + a bit larger network block
devices at all the common cloud providers have considerably higher
bandwidth. Even a raid5x of 4 spinning disks can deliver > 500MB/s.

And plenty of even the smaller instances at many providers have >
5gbit/s network. At the upper end it's way more than that.

Greetings,

Andres Freund