From: | Alfred Perlstein <bright(at)wintelcom(dot)net> |
---|---|
To: | tc lewis <tcl(at)bunzy(dot)net> |
Cc: | adb <adb(at)Beast(dot)COM>, Shaw Terwilliger <sterwill(at)sourcegear(dot)com>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: Re: Re: PostgreSQL over NFS? |
Date: | 2001-02-10 01:11:47 |
Message-ID: | 20010209171147.E3274@fw.wintelcom.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
* tc lewis <tcl(at)bunzy(dot)net> [010209 17:08] wrote:
>
> > >
> > > how's that? storage arrays will require you to use a filesystem that your
> > > operating system supports. or do some storage array companies provide
> > > their own filesystems and drivers for misc systems? same thing i guess.
> >
> > It has a persistant write cache, meaning it can ACK a write as being
> > sync'd to disk even if it's only been copied into the RAID's memory
> > cache (because it's persistant). Same thing that netapp does.
> >
>
> hmm, i suppose. i guess the filesystem wouldn't need to understand the
> logic of the underlying device. not sure. i don't think i've ever been
> presented with a storage array that does what you're saying (not saying
> they don't exist).
All it does is present itself as a disk that super fast at writing
as long as the cache doesn't fill up. I have an older card:
amr0: <AMI MegaRAID> port 0x1080-0x10ff irq 18 at device 11.0 on pci0
amr0: firmware UF80 bios 1.61 128MB memory
That has a slot for battery backup of the RAID cache. If the
machine was to reboot, the raid would sync the cached data to
disk sometime after coming back up.
--
-Alfred Perlstein - [bright(at)wintelcom(dot)net|alfred(at)freebsd(dot)org]
"I have the heart of a child; I keep it in a jar on my desk."
From | Date | Subject | |
---|---|---|---|
Next Message | George Johnson | 2001-02-10 01:16:25 | two cents |
Previous Message | tc lewis | 2001-02-10 01:08:10 | Re: Re: Re: PostgreSQL over NFS? |