Re: One process per session lack of sharing

From: AMatveev(at)bitec(dot)ru
To: Greg Stark <stark(at)mit(dot)edu>
Cc: Dave Cramer <pg(at)fastcrypt(dot)com>, Vladimir Sitnikov <sitnikov(dot)vladimir(at)gmail(dot)com>, Craig Ringer <craig(at)2ndquadrant(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: One process per session lack of sharing
Date: 2016-07-18 15:21:18
Message-ID: 486401842.20160718182118@bitec.ru
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi

> In other words, there's no theoretical reason you couldn't have adapt
> a JVM to create a large shared memory segment using mmap or SysV
I think even if I was the leader in OS development, I could not
correctly answer your question.
So just let discuss.
Ok, I agree with you that there is no " theoretical reason "
But in practice I think the main reason that OS(VM) developers
implement this things differently.

>there's no theoretical reason you couldn't
Why does Os developers make threads?
If there is no reason the use of thread just waste?
Why do most(any) common web server or balancer use thread?
May be they are bad theoretical?
and so on

But to be more constructive.
I just don't know how to make between process things that we can easily do
between threads, in most os.

I don't know how to share mutable execution code.
so i just cant imagine how to implement
http://docs.oracle.com/javase/7/docs/technotes/guides/vm/multiple-language-support.html#invokedynamic
in optimal way.

I don't understand why mutex has overhead compare to critical section
for windows.
http://stackoverflow.com/questions/800383/what-is-the-difference-between-mutex-and-critical-section

And so on.

So I think as long as process and thread have different function in OS,
use process like thread will have overhead in practice.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message AMatveev 2016-07-18 16:26:30 Re: One process per session lack of sharing
Previous Message Tom Lane 2016-07-18 15:13:02 Re: Regression tests vs existing users in an installation