Re: Subtransactions + a long-running transaction leading to performance degradation on standbys

Поиск
Список
Период
Сортировка
От Nikolay Samokhvalov
Тема Re: Subtransactions + a long-running transaction leading to performance degradation on standbys
Дата
Msg-id CAM527d8XCfmyVyaHbzJALM_m1g8Ysua7Vr8SGdYMrHtB4mpKBw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Subtransactions + a long-running transaction leading to performance degradation on standbys  (Andrey Borodin <x4mmm@yandex-team.ru>)
Список pgsql-hackers
Дерево обсуждения

On Thu, Aug 19, 2021 at 10:04 PM Andrey Borodin <x4mmm@yandex-team.ru> wrote:
I just want to note, that on your screenshot unpatched version runs 400K tps, while patched runs 280K tps. I see the dates are different and this effect is not observed in  [0]. Probably, you run tests on different machines.

Indeed, patched Postgres I was testing on 2x smaller EC2 instances, it is documented in the GitLab issue. But I added an additional note here: https://gitlab.com/postgres-ai/postgresql-consulting/tests-and-benchmarks/-/issues/21#note_655731979
 
While your experiments clearly shows that patch can save DB from degradation under pathological workload it would be great to ensure patch does not incur penalty on normal workload.

Makes sense. I'll try to find time to make a direct comparison.

I documented all the steps in detail in the GitLab issue, so anyone interested can reproduce it and explore the problem at different angles.

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Andrey Borodin
Дата:
Сообщение: Re: Subtransactions + a long-running transaction leading to performance degradation on standbys
Следующее
От: Michael Paquier
Дата:
Сообщение: Re: Two patches to speed up pg_rewind.