Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 21 Nov 1999 22:50:00 +0200
From:      Iani Brankov <ian@bulinfo.net>
To:        Freebsd-questions <freebsd-questions@freebsd.org>
Subject:   Do I have NO FUTURE with vinum?
Message-ID:  <38385AF8.731806A9@bulinfo.net>

next in thread | raw e-mail | index | archive | help
Hi everybody,

I have a big headaches with this 'evil'.
I configured a two disk striped volume, but it performs much slower
than I expected.

The read performance - ~20MB/s is ok, but in writing it's ~4MB/s. Both
the separate disks are much faster than the stripped volume,
configured onto them.

My machine is PII/350MHz with an Intel BX board, 64MB RAM, Intel
EtherExpress Pro/100B, Adaptec U2W SCSI controller with IBM DDRS 9.1GB
& IBM DNES 9.1GB LVD disks working in LVD mode and a Quantum Fireball
SE 4.3BG single ended narrow scsi disk used for 3.3 FreeBSD cvsupped a
day ago.

I think the SCSI chain is ok (termination, length, etc)

I also tried the performance configuring a 'one disk concatenated'
volume on the first hard drive, but the result was terrible again
(below 50% of the standalone disk performance).

I noticed when I write onto the raw vinum device it performs pretty
well, whereas the cooked one is good in reading, but very slow in
writing.

I made so many tests in different ways, so I cannot believe it's
related to a configuration mistake. Anyway here's the vinum.cfg:


drive d1 device /dev/da1h
drive d2 device /dev/da2h
volume double
	plex org striped 256k
		sd length 8700mb drive d1
		sd length 8700mb drive d2


The hardware configuration is very popular, so I hope someone has
already solved this problem, if it's related to the hardware.

Thanks in advance,
iani


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?38385AF8.731806A9>