Re: Do you know the reason for increased max latency due to xlog scaling?

Поиск
Список
Период
Сортировка
От Jeff Janes
Тема Re: Do you know the reason for increased max latency due to xlog scaling?
Дата
Msg-id CAMkU=1znu42ODAFMkjdmJxgEM7_Ggy1kWKgS2Tr6T_UhSVeq=w@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Do you know the reason for increased max latency due to xlog scaling?  ("MauMau" <maumau307@gmail.com>)
Ответы Re: Do you know the reason for increased max latency due to xlog scaling?  (Heikki Linnakangas <hlinnakangas@vmware.com>)
Список pgsql-hackers
Дерево обсуждения
On Tue, Feb 18, 2014 at 3:49 AM, MauMau <maumau307@gmail.com> wrote:
From: "Andres Freund" <andres@2ndquadrant.com>
On 2014-02-18 01:35:52 +0900, MauMau wrote:
For example, please see the max latencies of test set 2 (PG 9.3) and test
set 4 (xlog scaling with padding).  They are 207.359 and 1219.422
respectively.  The throughput is of course greatly improved, but I think the
response time should not be sacrificed as much as possible.  There are some
users who are sensitive to max latency, such as stock exchange and online
games.

You need to compare both at the same throughput to have any meaningful
comparison.

I'm sorry for my lack of understanding, but could you tell me why you think so?  When the user upgrades to 9.4 and runs the same workload, he would experience vastly increased max latency

The tests shown have not tested that.  The test is not running the same workload on 9.4, but rather a vastly higher workload.  If we were to throttle the workload in 9.4 (using pgbench's new -R, for example) to the same level it was in 9.3, we probably would not see the max latency increase.  But that was not tested, so we don't know for sure.

 
--- or in other words, greater variance in response times.  With my simple understanding, that sounds like a problem for response-sensitive users.

If you need the throughput provided by 9.4, then using 9.3 gets lower variance simply be refusing to do 80% of the assigned work.  If you don't need the throughput provided by 9.4, then you probably have some natural throttling in place.
 
If you want a real-world like test, you might try to crank up the -c and -j to the limit in 9.3 in a vain effort to match 9.4's performance, and see what that does to max latency.  (After all, that is what a naive web app is likely to do--continue to make more and more connections as requests come in faster than they can finish.)

Cheers,

Jeff

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: OS X and ossp-uuid, next chapter
Следующее
От: "MauMau"
Дата:
Сообщение: Re: [bug fix] postgres.exe fails to start on Windows Server 2012 due to ASLR