Are hardware vendors starting to bail on FreeBSD ... ?
Ted Mittelstaedt
tedm at toybox.placo.com
Thu Jul 27 15:30:25 UTC 2006
----- Original Message -----
From: "Nikolas Britton" <nikolas.britton at gmail.com>
To: "Ted Mittelstaedt" <tedm at toybox.placo.com>
Cc: "Greg Barniskis" <nalists at scls.lib.wi.us>; <jerrymc at clunix.cl.msu.edu>;
<danial_thom at yahoo.com>; <freebsd-questions at freebsd.org>; "Nick Withers"
<nick at nickwithers.com>
Sent: Wednesday, July 26, 2006 11:38 PM
Subject: Re: Are hardware vendors starting to bail on FreeBSD ... ?
> On 7/27/06, Ted Mittelstaedt <tedm at toybox.placo.com> wrote:
> >
> > ----- Original Message -----
> > From: "Greg Barniskis" <nalists at scls.lib.wi.us>
> > To: "Ted Mittelstaedt" <tedm at toybox.placo.com>
> > Cc: <jerrymc at clunix.cl.msu.edu>; <danial_thom at yahoo.com>;
> > <freebsd-questions at freebsd.org>; "Nick Withers" <nick at nickwithers.com>
> > Sent: Monday, July 24, 2006 11:59 AM
> > Subject: Re: Are hardware vendors starting to bail on FreeBSD ... ?
> >
> >
> > > Ted Mittelstaedt wrote:
> > > > ----- Original Message -----
> > > > From: "Danial Thom" <danial_thom at yahoo.com>
> > > > To: "Greg Barniskis" <nalists at scls.lib.wi.us>; "Nick Withers"
> > > > <nick at nickwithers.com>
> > > > Cc: <jerrymc at clunix.cl.msu.edu>; <danial_thom at yahoo.com>;
> > > > <freebsd-questions at freebsd.org>
> > > > Sent: Thursday, July 13, 2006 11:10 AM
> > > > Subject: Re: Are hardware vendors starting to bail on FreeBSD ... ?
> > > >
> > > >
> > > >> Burying your head in the sand is a common method
> > > >> used by stupid people that have no answer to the
> > > >> truth. I don't blame you; you guys don't want
> > > >> your employers to know that you've wasted man
> > > >> 1000s of their dollars because you don't know the
> > > >> performance characteristics of the hardware
> > > >> you've recommended. It must be thoroughly
> > > >> embarrassing.
> > > [snip]
> > >
> > > > I do agree with Danial that most USERS on this list are
> > > > burying their heads in the sand on this issue. But I will
> > > > point out that there isn't really any reason they shouldn't
> > > > be. What the market wants is features, not speed. And
> > > > that is what the FreeBSD developers are working on.
> > >
> > > Features over speed is generally the right equation, yes.
> > >
> > > But I think you're being too generous to Danial. The quote of his
> > > above was in direct response to my assertion that many people refuse
> > > to listen to him because he frequently engages in cheap demagogy[1].
> > >
> >
> > He does, but he is also right on this performance point. The truth
> > can always be wrapped more palatably, but I think one of the differences
> > between a system administrator and a user is that a user can't deal with
> > the truth unless it's spoon fed in the nursery, an administrator should
> > be approaching it as a professional, which means ignoring the
> > irrelevant cheap demagogery and ignoring their own preconceptions of
> > how things are "supposed" to work, and paying attention to the kernels
of
> > truth.
> >
> > I have to sort through giant piles of horseshit every time I look at the
> > latest Cisco sales and marketing dreck, to find out what might be
> > important in one of their new products, this isn't any different. And
> > frankly I find the saccherine cloying marketingspeak to be far more
> > disgusting and offensive then the lame kindergarden flames that
> > Danial has so far been able to come up with.
> >
> > > His response? Another whole boatload of cheap demagogy, questioning
> > > the intelligence, aptitude and moral character of anyone who doesn't
> > > listen to him, by way of accusations that are wholly unsupported by
> > > facts. I could probably rest my case right there, but I think his
> > > perception (and yours) that people are not receptive to claims of
> > > FreeBSD performance problems is quite simply false.
> > >
> > > Every time a performance question is brought up, I see a flurry of
> > > calls for clarification and for the formulation of repeatable tests
> > > which are generally agreed to be an accurate gauge of the problem.
> >
> > Calling for testing is pretty much a way of excusing the claim. People
> > including Danial, have done the testing in the past, posted the results,
> > then had armchair quarterbacks pick apart the test methodology claiming
> > the tests were done wrong, thus irrelevant. So why even bother doing
> > it anymore.
> >
> > But, you asked for it, you got it:
> >
> > Machine #1: Compaq 1600R, FBSD 6.1 Pentium 3 550Mhz
> >
> > freebsd-cvs# dmesg
> > Copyright (c) 1992-2006 The FreeBSD Project.
> > Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
> > The Regents of the University of California. All rights
reserved.
> > FreeBSD 6.1-RELEASE #0: Thu Jun 1 17:23:18 PDT 2006
> >
tedm at freebsd-cvs.ipinc.net:/usr/src/sys/i386/compile/GENERICNOUSBNOFIRE
> > Timecounter "i8254" frequency 1193182 Hz quality 0
> > CPU: Pentium III/Pentium III Xeon/Celeron (548.54-MHz 686-class CPU)
> > Origin = "GenuineIntel" Id = 0x673 Stepping = 3
> >
> >
Features=0x383fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,
> > CMOV,PAT,PSE36,MMX,FXSR,SSE>
> > real memory = 671088640 (640 MB)
> > avail memory = 647458816 (617 MB)
> > MPTable: <COMPAQ PROLIANT >
> > ioapic0: Changing APIC ID to 8
> > ioapic0: Assuming intbase of 0
> > ioapic0 <Version 1.1> irqs 0-34 on motherboard
> > kbd1 at kbdmux0
> > cpu0 on motherboard
> > pcib0: <MPTable Host-PCI bridge> pcibus 0 on motherboard
> > pci0: <PCI bus> on pcib0
> > pci0: <display, VGA> at device 11.0 (no driver attached)
> > pcib1: <MPTable PCI-PCI bridge> at device 13.0 on pci0
> > pci1: <PCI bus> on pcib1
> > tl0: <Compaq Netelligent 10/100 Proliant> port 0x3800-0x380f irq 30 at
> > device 7.0 on pci1
> > miibus0: <MII bus> on tl0
> > nsphy0: <DP83840 10/100 media interface> on miibus0
> > nsphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
> > tlphy0: <ThunderLAN 10baseT media interface> on miibus0
> > tlphy0: 10base2/BNC, 10base5/AUI
> > tl0: Ethernet address: 00:50:8b:f1:82:17
> > sym0: <875> port 0x3000-0x30ff mem
> > 0xc6ffdf00-0xc6ffdfff,0xc6fff000-0xc6ffffff irq 23 at device 9.0 on pci1
> > sym0: No NVRAM, ID 7, Fast-20, SE, parity checking
> > sym0: [GIANT-LOCKED]
> > sym1: <875> port 0x3400-0x34ff mem
> > 0xc6ffde00-0xc6ffdeff,0xc6ffe000-0xc6ffefff irq 22 at device 9.1 on pci1
> > sym1: No NVRAM, ID 7, Fast-20, SE, parity checking
> > sym1: [GIANT-LOCKED]
> > pci0: <base peripheral> at device 14.0 (no driver attached)
> > ida0: <Compaq Smart Array 431 controller> port 0x2000-0x20ff mem
> > 0xc6efe000-0xc6efefff irq 20 at device 16.0 on pci0
> > ida0: [GIANT-LOCKED]
> > ida0: drives=1 firm_rev=1.22
> > idad0: <Compaq Logical Drive> on ida0
> > idad0: 34719MB (71106240 sectors), blocksize=512
> > pci0: <serial bus, USB> at device 18.0 (no driver attached)
> > isab0: <PCI-ISA bridge> at device 20.0 on pci0
> > isa0: <ISA bus> on isab0
> > atapci0: <Intel PIIX4 UDMA33 controller> port
> > 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xf100-0xf10f at device 20.1 on pci0
> > ata0: <ATA channel 0> on atapci0
> > ata1: <ATA channel 1> on atapci0
> > pci0: <serial bus, USB> at device 20.2 (no driver attached)
> > pci0: <bridge> at device 20.3 (no driver attached)
> > eisa0: <EISA bus> on motherboard
> > mainboard0: <CPQ0689 (System Board)> on eisa0 slot 0
> > pmtimer0 on isa0
> > orm0: <ISA Option ROMs> at iomem
> > 0xc0000-0xc7fff,0xc8000-0xcbfff,0xe8000-0xedfff,0xee000-0xeffff on isa0
> > atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
> > atkbd0: <AT Keyboard> irq 1 on atkbdc0
> > kbd0 at atkbd0
> > atkbd0: [GIANT-LOCKED]
> > psm0: <PS/2 Mouse> irq 12 on atkbdc0
> > psm0: [GIANT-LOCKED]
> > psm0: model Generic PS/2 mouse, device ID 0
> > fdc0: <Enhanced floppy controller> at port 0x3f0-0x3f5,0x3f7 irq 6 drq 2
on
> > isa0
> > fdc0: [FAST]
> > fd0: <1440-KB 3.5" drive> on fdc0 drive 0
> > ppc0: <Parallel port> at port 0x3bc-0x3c3 irq 7 on isa0
> > ppc0: Generic chipset (NIBBLE-only) in COMPATIBLE mode
> > ppbus0: <Parallel port bus> on ppc0
> > plip0: <PLIP network interface> on ppbus0
> > lpt0: <Printer> on ppbus0
> > lpt0: Interrupt-driven port
> > ppi0: <Parallel I/O> on ppbus0
> > sc0: <System console> at flags 0x100 on isa0
> > sc0: VGA <16 virtual consoles, flags=0x300>
> > sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0
> > sio0: type 16550A
> > sio1 at port 0x2f8-0x2ff irq 3 on isa0
> > sio1: type 16550A
> > vga0: <Generic ISA VGA> at port 0x3c0-0x3df iomem 0xa0000-0xbffff on
isa0
> > Timecounter "TSC" frequency 548543576 Hz quality 800
> > Timecounters tick every 1.000 msec
> > Waiting 5 seconds for SCSI devices to settle
> > acd0: CDROM <COMPAQ CDR-8435/0013> at ata0-master PIO4
> > sa0 at sym1 bus 0 target 6 lun 0
> > sa0: <COMPAQ SDX-400C 3.0B> Removable Sequential Access SCSI-2 device
> > sa0: 40.000MB/s transfers (20.000MHz, offset 7, 16bit)
> > Trying to mount root from ufs:/dev/idad0s1a
> > ipfw2 (+ipv6) initialized, divert loadable, rule-based forwarding
disabled,
> > default to deny, logging disabled
> > tl0: link state changed to UP
> > freebsd-cvs#
> > freebsd-cvs# pwd
> > /root
> > freebsd-cvs# bonnie
> > File './Bonnie.81888', size: 104857600
> > Writing with putc()...done
> > Rewriting...done
> > Writing intelligently...done
> > Reading with getc()...done
> > Reading intelligently...done
> > Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done...
> > -------Sequential Output-------- ---Sequential
> > Input-- --Random--
> > -Per Char- --Block--- -Rewrite-- -Per
> > Char- --Block--- --Seeks---
> > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU
/sec
> > %CPU
> > 100 7745 39.4 7748 12.0 8808 15.3 23303 98.3 151922 99.1
> > 12579.6 99.1
> > freebsd-cvs#
> >
> > Machine #2: Compaq 1600R, FBSD 4.11 Pentium 3 550Mhz
> >
> > billmax# dmesg
> > Copyright (c) 1992-2005 The FreeBSD Project.
> > Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
> > The Regents of the University of California. All rights
reserved.
> > FreeBSD 4.11-RELEASE #0: Mon Nov 14 11:25:24 PST 2005
> > tedm at billmax.ipinc.net:/usr/src/sys/compile/BILLMAX
> > Timecounter "i8254" frequency 1193182 Hz
> > CPU: Pentium II/Pentium II Xeon/Celeron (448.95-MHz 686-class CPU)
> > Origin = "GenuineIntel" Id = 0x652 Stepping = 2
> >
> >
Features=0x183fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,
> > CMOV,PAT,PSE36,MMX,FXSR>
> > real memory = 536870912 (524288K bytes)
> > avail memory = 518692864 (506536K bytes)
> > Changing APIC ID for IO APIC #0 from 0 to 8 on chip
> > Programming 35 pins in IOAPIC #0
> > IOAPIC #0 intpin 2 -> irq 0
> > FreeBSD/SMP: Multiprocessor motherboard: 2 CPUs
> > cpu0 (BSP): apic id: 1, version: 0x00040011, at 0xfee00000
> > cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000
> > io0 (APIC): apic id: 8, version: 0x00220011, at 0xfec00000
> > Preloaded elf kernel "kernel" at 0xc039c000.
> > Pentium Pro MTRR support enabled
> > md0: Malloc disk
> > npx0: <math processor> on motherboard
> > npx0: INT 16 interface
> > pcib0: <Intel 82443BX host to PCI bridge (AGP disabled)> on motherboard
> > pci0: <PCI bus> on pcib0
> > pci0: <Cirrus Logic GD5446 SVGA controller> at 11.0
> > pcib1: <DEC 21150 PCI-PCI bridge> at device 13.0 on pci0
> > pci1: <PCI bus> on pcib1
> > tl0: <Compaq Netelligent 10/100 Proliant> port 0x2800-0x280f mem
> > 0xc6ef9df0-0xc6ef9dff irq 9 at device 7.0 on pci1
> > tl0: Ethernet address: 00:08:c7:9f:92:77
> > miibus0: <MII bus> on tl0
> > nsphy0: <DP83840 10/100 media interface> on miibus0
> > nsphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
> > tlphy0: <ThunderLAN 10baseT media interface> on miibus0
> > tlphy0: 10base2/BNC, 10base5/AUI
> > sym0: <875> port 0x2000-0x20ff mem
> > 0xc6efb000-0xc6efbfff,0xc6ef9f00-0xc6ef9fff irq 10 at device 9.0 on pci1
> > sym0: No NVRAM, ID 7, Fast-20, SE, parity checking
> > sym1: <875> port 0x2400-0x24ff mem
> > 0xc6efa000-0xc6efafff,0xc6ef9e00-0xc6ef9eff irq 11 at device 9.1 on pci1
> > sym1: No NVRAM, ID 7, Fast-20, SE, parity checking
> > pci1: <Matrox MGA Millennium 2064W graphics accelerator> at 13.0 irq 5
> > pci0: <unknown card> (vendor=0x0e11, dev=0xa0f0) at 14.0
> > pcib2: <IBM 82351 PCI-PCI bridge> at device 18.0 on pci0
> > pci2: <PCI bus> on pcib2
> > ida0: <Compaq SMART-2SL array controller> port 0x3000-0x30ff mem
> > 0xb8000000-0xbfffffff,0xc6ffff00-0xc6ffffff irq 15 at device 0.0 on pci2
> > ida0: drives=1 firm_rev=4.44
> > idad0: <Compaq Logical Drive> on ida0
> > idad0: 26029MB (53309280 sectors), blocksize=512
> > isab0: <Intel 82371AB PCI to ISA bridge> at device 20.0 on pci0
> > isa0: <ISA bus> on isab0
> > atapci0: <Intel PIIX4 ATA33 controller> port 0xf100-0xf10f at device
20.1 on
> > pci0
> > ata0: at 0x1f0 irq 14 on atapci0
> > ata1: at 0x170 irq 15 on atapci0
> > uhci0: <Intel 82371AB/EB (PIIX4) USB controller> irq 0 at device 20.2 on
> > pci0
> > uhci0: Could not map ports
> > device_probe_and_attach: uhci0 attach returned 6
> > piix0: <Intel 82371AB Power management controller> at device 20.3 on
pci0
> > eisa0: <EISA bus> on motherboard
> > mainboard0: <CPQ0689 (System Board)> on eisa0 slot 0
> > orm0: <Option ROMs> at iomem
> > 0xc0000-0xc7fff,0xc8000-0xcbfff,0xe8000-0xedfff,0xee000-0xeffff on isa0
> > pmtimer0 on isa0
> > fdc0: <NEC 72065B or clone> at port 0x3f0-0x3f5,0x3f7 irq 6 drq 2 on
isa0
> > fdc0: FIFO enabled, 8 bytes threshold
> > fd0: <1440-KB 3.5" drive> on fdc0 drive 0
> > atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
> > atkbd0: <AT Keyboard> flags 0x1 irq 1 on atkbdc0
> > kbd0 at atkbd0
> > psm0: <PS/2 Mouse> irq 12 on atkbdc0
> > psm0: model Generic PS/2 mouse, device ID 0
> > vga0: <Generic ISA VGA> at port 0x3c0-0x3df iomem 0xa0000-0xbffff on
isa0
> > sc0: <System console> at flags 0x100 on isa0
> > sc0: VGA <16 virtual consoles, flags=0x300>
> > sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0
> > sio0: type 16550A
> > sio1 at port 0x2f8-0x2ff irq 3 on isa0
> > sio1: type 16550A
> > APIC_IO: Testing 8254 interrupt delivery
> > APIC_IO: routing 8254 via IOAPIC #0 intpin 2
> > SMP: AP CPU #1 Launched!
> > acd0: CDROM <CD-ROM CDU701-Q> at ata0-master PIO4
> > Waiting 15 seconds for SCSI devices to settle
> > sa0 at sym0 bus 0 target 6 lun 0
> > sa0: <HP C1537A L708> Removable Sequential Access SCSI-2 device
> > sa0: 10.000MB/s transfers (10.000MHz, offset 16)
> > Mounting root from ufs:/dev/idad0s1a
> > IP packet filtering initialized, divert disabled, rule-based forwarding
> > enabled, default to deny, logging disabled
> > (sa0:sym0:0:6:0): WRITE FILEMARKS. CDB: 10 0 0 0 2 0
> > (sa0:sym0:0:6:0): Deferred Error: MEDIUM ERROR asc:3b,0
> > (sa0:sym0:0:6:0): Sequential positioning error
> > billmax#
> > billmax# bonnie
> > File './Bonnie.12377', size: 104857600
> > Writing with putc()...done
> > Rewriting...done
> > Writing intelligently...done
> > Reading with getc()...done
> > Reading intelligently...done
> > Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done...
> > -------Sequential Output-------- ---Sequential
> > Input-- --Random--
> > -Per Char- --Block--- -Rewrite-- -Per
> > Char- --Block--- --Seeks---
> > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU
/sec
> > %CPU
> > 100 6191 27.7 6202 8.6 7269 13.6 14247 100.0 122401 99.8
> > 12267.0 199.0
> > billmax#
> >
> > Notice on the FASTER machine, running FreeBSD 6.1, it runs CPU at 39.4
> > percent. The SLOWER
> > machine, running FreeBSD 4.11, runs CPU at 27%. Granted, disk I/O is a
bit
> > faster on the newer version
> > of FreeBSD. But of course, it should be - the disk driver and OS is
newer.
> > The dirty little secret is that
> > while you might get that higher throughput speed on the disk, it takes a
LOT
> > higher CPU percentage,
> > EVEN ON a machine that's 100Mhz faster.
> >
> > Sure, if you got gigahertz to play with - so what? But, this matters a
> > great deal on older hardware
> > where you DON'T got that.
> >
> > Now, granted this is nothing more than seat of the pants testing, it
isn't a
> > controlled test.
> > But it is very typical of the results that people get when they try
> > performance testing.
> >
>
> Your results on CPU usage are useless. Besides the difference in CPU
> speed
explain how a faster CPU speed would increase CPU usage.
> there is at least one more major problem you missed, the 4.11
> system is SMP and the 6.1 system is not. anyways...
>
bonnie isn't multithreaded, only one CPU is going to have it. But
it is true the systems are different. That is why I said the testing
wasn't going to be more useful than seat-of-the-pants.
> Here's what I'd like to know. What's your problem Danial? After two
> years of trolling I still don't know what the root of your complaint
> is. I'm willing to discuss this (as I've always been) provided you,
> and everyone else, can stay civil. I would also like to run some of
> these tests you keep referring to.
>
He's referred to network throughput tests. As in you have a FBSD system
with 2 nics, and you pass a lot of data from 1 nic to the other. Not a very
easy test for most people to duplicate, I'm afraid.
Ted
More information about the freebsd-questions
mailing list