Re: POC: Cache data in GetSnapshotData()

Поиск
Список
Период
Сортировка
От Mithun Cy
Тема Re: POC: Cache data in GetSnapshotData()
Дата
Msg-id CAD__OuhygfLox2grcL4=jyLjGMsLmPiWqCi05PwrbOoqzqNyYw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: POC: Cache data in GetSnapshotData()  (Robert Haas <robertmhaas@gmail.com>)
Ответы Re: POC: Cache data in GetSnapshotData()  (Amit Kapila <amit.kapila16@gmail.com>)
Список pgsql-hackers


On Thu, Mar 3, 2016 at 11:50 PM, Robert Haas <robertmhaas@gmail.com> wrote:
>What if you apply both this and Amit's clog control log patch(es)?  Maybe the combination of the two helps substantially more than either >one alone.


I did the above tests along with Amit's clog patch. Machine :8 socket, 64 core. 128 hyperthread.

clientsBASEONLY CLOG CHANGES% IncreaseONLY SAVE SNAPSHOT% IncreaseCLOG CHANGES + SAVE SNAPSHOT% Increase
6429247.65803430855.7288355.498118171129254.5321860.023503256232691.83277611.7758992463
8831214.30539133313.3930956.724761860632109.2486092.867093170235433.65520313.5173592978
12830896.67394934015.36215210.0939285832******34947.29635513.110221549
25627183.78092131192.89543714.7481857938******32873.78273520.9316056164

With clog changes, perf of caching the snapshot patch improves.

--
Thanks and Regards
Mithun C Y

В списке pgsql-hackers по дате отправления:

Предыдущее
От:
Дата:
Сообщение: Re: [PROPOSAL] VACUUM Progress Checker.
Следующее
От: Thomas Munro
Дата:
Сообщение: Re: Proposal: "Causal reads" mode for load balancing reads without stale data