[Koha-bugs] [Bug 19532] Recalls for Koha

[bugzilla-daemon at bugs.koha-community.org]

https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19532

Kyle M Hall <kyle at bywatersolutions.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kyle at bywatersolutions.com

--- Comment #41 from Kyle M Hall <kyle at bywatersolutions.com> ---
> We did, but the reason old_issues exists, is because keeping it one massive
> table becomes really slow quite fast.

I think maybe we need to check our assumptions. We may have be running on best
practices that are out of date. I think that happens to us all from time to
time.

MySQL is capable of dealing with tens or even hundreds of millions of rows very
speedily, as long as the correct indexes are in place.

For example, I picked out a database to test with from a Koha consortium. They
have  1,456,666 old_issues rows, and 50122 issues rows.

Searching on a column with no index ( such as issuedate ) returned a single
record from almost 1.5 million records in 1.04 seconds. Not terrible, but
definitely not great. However, searching for the same row using an indexed
column took less time than mysql could measure! It reported back taking 0.00
seconds ; )

I've done some digging and it appears that many people are using mysql for
tables with tens to hundreds of millions of rows with no adverse effects.

Perhaps that was an issue during the early years of Koha, when MySQL was much
less mature, but I don't think it is anymore.

-- 
You are receiving this mail because:
You are watching all bug changes.