Serializable wrong?

From: Joshua Drake <jd(at)commandprompt(dot)com>
To: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Serializable wrong?
Date: 2020-06-12 17:58:25
Message-ID: CAJvJg-QdcQ7raGRrgmOEgApWHC=1cqd0G2uOvh=qSHP0k6gDsg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

-Hackers,

I came across this today [1], "
3 Results

In most respects, PostgreSQL behaved as expected: both read uncommitted and
read committed prevent write skew and aborted reads. We observed no
internal consistency violations. However, we have two surprising results to
report. The first is that PostgreSQL’s “repeatable read” is weaker than
repeatable read, at least as defined by Berenson, Adya, Bailis, et al. This
is not necessarily wrong: the ANSI SQL standard is ambiguous. The second
result, which is definitely wrong, is that PostgreSQL’s “serializable”
isolation level isn’t serializable: it allows G2-item during normal
operation. "

Thanks!

JD

1. https://jepsen.io/analyses/postgresql-12.3

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Gierth 2020-06-12 18:02:37 Re: Infinities in type numeric
Previous Message Robert Haas 2020-06-12 17:58:12 Re: Parallel Seq Scan vs kernel read ahead