From owner-freebsd-current Mon Dec 9 00:25:14 1996 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id AAA17236 for current-outgoing; Mon, 9 Dec 1996 00:25:14 -0800 (PST) Received: from btp1da.phy.uni-bayreuth.de (btp1da.phy.uni-bayreuth.de [132.180.20.32]) by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id AAA17218 for ; Mon, 9 Dec 1996 00:25:07 -0800 (PST) Received: (from root@localhost) by btp1da.phy.uni-bayreuth.de (8.8.4/8.7.3) id JAA27943; Mon, 9 Dec 1996 09:24:18 +0100 (CET) From: Werner Griessl Message-Id: <199612090824.JAA27943@btp1da.phy.uni-bayreuth.de> Subject: Re: rdump slow In-Reply-To: <199612072019.VAA22287@uriah.heep.sax.de> from J Wunsch at "Dec 7, 96 09:19:54 pm" To: joerg_wunsch@uriah.heep.sax.de Date: Mon, 9 Dec 1996 09:24:18 +0100 (CET) Cc: current@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > As Werner Griessl wrote: > > > rdump in current is very slow, should be ~5 times faster : > > > > Wed Dec 4 09:57:01 1996 start btp1da:system_save to btp1x5:/dev/nrmt0h > > Wed Dec 4 09:57:01 1996 rewinding tape > > Wed Dec 4 09:57:07 1996 #1 rdump / ... > > > DUMP: DUMP: 21409 tape blocks on 1 volumes(s) > > > > DUMP: finished in 430 seconds, throughput 49 KBytes/sec > > ^^^^^^^^^^^^^ !!!!! > > I can't confirm this. What's your blocksize? What system is the > remote TCP peer? Is the tape streaming? > Blocksize is the default (10), remote system is a DEC-alpha 3000/600, tape is a HP-DAT 35480 with local transfer-rate ~250 kb/sec . > That's what i get here: > > j@uriah 1523% /sbin/dump 0Bbf 5000000 32 localhost:/dev/rst0 / > ... > DUMP: DUMP: 20154 tape blocks on 1 volumes(s) > DUMP: finished in 53 seconds, throughput 380 KBytes/sec > ... > j@uriah 1524% /sbin/dump 0Bbf 5000000 32 /dev/rst0 / > ... > DUMP: DUMP: 20154 tape blocks on 1 volumes(s) > DUMP: finished in 53 seconds, throughput 380 KBytes/sec > > So of course, it's been the loopback device, but as long as the > Ethernet card in question can handle 380 KB/s (which is not very > much), the limiting factor is obviously the tape here. (It's a > QIC-2.5GB w/ compression in a Tandberg drive. The 380 KB/s is a > normal rate there.) > > -- > cheers, J"org > > joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE > Never trust an operating system you don't have sources for. ;-) >