From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Denis Perchine <dyp(at)perchine(dot)com> |
Cc: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Performance monitor |
Date: | 2001-03-13 15:19:20 |
Message-ID: | 19312.984496760@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Denis Perchine <dyp(at)perchine(dot)com> writes:
>>> Small question... Will it work in console? Or it will be X only?
>>
>> It will be tck/tk, so I guess X only.
> That's bad.
tcl/tk is cross-platform; there's no reason that a tcl-coded
performance monitor client couldn't run on Windows or Mac.
The real problem with the ps-based implementation that Bruce is
proposing is that it cannot work remotely at all, because there's
no way to get the ps data from another machine (unless you're
oldfashioned/foolish enough to be running a finger server that
allows remote ps). This I think is the key reason why we'll
ultimately want to forget about ps and go to a shared-memory-based
arrangement for performance info. That could support a client/server
architecture where the server is a backend process (or perhaps a
not-quite-backend process, but anyway attached to shared memory)
and the client is communicating with it over TCP.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Thomas Lockhart | 2001-03-13 15:31:02 | Re: PostgreSQL on multi-CPU systems |
Previous Message | Denis Perchine | 2001-03-13 15:15:17 | Re: problem with fe/be protocol and large objects |