[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openbsd-advocacy
Subject:    Soft Updates are cool!
From:       Nick Holland <nick () holland-consulting ! net>
Date:       2002-06-01 23:45:54
[Download RAW message or body]

I like to time(1) the unpacking of ports.tar.gz when I set up a new
system.  Not sure why, though it gives me an indication of the speed
of the disk system.  

Just recently, I realized that a lot of what I do with my systems
around here could REALLY be improved by using Soft Updates.  I don't
know why I never did this before...but if you rm -r /usr/obj (or
ANYTHING) on a Mac68k a few times, perhaps, you really start looking
for things to improve life...

I just set up a machine I'm planning to use as a terminal server.  Not
much, a P100, 80M RAM, 1.2G IDE drive, etc. (dmesg below), just enough
to put an 8 port serial board in to monitor a bunch of low-usage
machines.  These machines are solid, but slow...no PCIIDE support,
uses wdc(4) as its driver, not the kind of thing you would expect to
win races with.

I run the following:
  # time tar xzf ports.tar.gz
  1031.31s real    5.48s user     86.26s system
  #
This is more-or-less typical, it seems.  On a fast system, I might get
half that...but keep in mind, no dma, no udma, and old, slow drive. 
(On a really slow system, I might get double that or more...  On a
Mac68k, I get far less, but that is because Mac68ks cheat, and stop
the clock 8-)


Now...delete the ports directory I just made, enable soft updates, and
try again:
  # time tar xzf ports.tar.gz
  368.01s real     5.03s user     48.46s system
  #

That's a big change.  Unpacking (or deleting!) ports involves
thousands of tiny files, and this is what Soft Updates is very good
at.
  
Now, I would be remiss if I didn't point out that Softdeps aren't for
everyone, they don't work on Sun4c or Sun4 systems (kernel memory
issues), and they are not considered appropriate by some very
knowledgeable people for mail queue partitions, but when you can use
softdeps, they can give you a huge performance boost.

Read more about 'em here:
   http://www.openbsd.org/faq/faq14.html#SoftUpdates

If you haven't tried Soft Updates, you owe it to yourself to do so.
8-)

Nick.
-- 
http://www.holland-consulting.net

OpenBSD 3.1-current (GENERIC) #193: Fri May 31 23:39:39 MDT 2002
    deraadt@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC
cpu0: F00F bug workaround installed
cpu0: Intel Pentium (P54C) ("GenuineIntel" 586-class) 100 MHz
cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8
real mem  = 83472384 (81516K)
avail mem = 71655424 (69976K)
using 1044 buffers containing 4276224 bytes (4176K) of memory
mainbus0 (root)
bios0 at mainbus0: AT/286+(00) BIOS, date 05/12/95, BIOS32 rev. 0 @
0xf143c
pcibios0 at bios0: rev. 2.1 @ 0xf0000/0x10000
pcibios0: PCI BIOS has 5 Interrupt Routing table entries
pcibios0: no compatible PCI ICU found
pcibios0: PCI bus #0 is the last bus
bios0: ROM list: 0xc0000/0x8000 0xed000/0x1000
pci0 at mainbus0 bus 0: configuration mode 2 (bios)
pchb0 at pci0 dev 0 function 0 "Intel 82434LX/NX PCI/Cache/DRAM" rev
0x11
"PC Technology RZ1000" rev 0x01 at pci0 dev 1 function 0 not
configured
pcib0 at pci0 dev 2 function 0 "Intel 82378IB PCI-ISA" rev 0x43
vga1 at pci0 dev 3 function 0 "ATI Mach64 GX" rev 0x01
wsdisplay0 at vga1: console (80x25, vt100 emulation)
wsdisplay0: screen 1-5 added (80x25, vt100 emulation)
fxp0 at pci0 dev 12 function 0 "Intel 82557" rev 0x02: irq 11, address
00:a0:c9:5f:f6:98
inphy0 at fxp0 phy 1: i82555 10/100 media interface, rev. 0
isa0 at pcib0
isadma0 at isa0
pckbc0 at isa0 port 0x60/5
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard, using wsdisplay0
pms0 at pckbc0 (aux slot)
pckbc0: using irq 12 for aux slot
wsmouse0 at pms0 mux 0
wdc0 at isa0 port 0x1f0/8 irq 14
wd0 at wdc0 channel 0 drive 0: <QUANTUM FIREBALL1280A>
wd0: 8-sector PIO, LBA, 1222MB, 2484 cyl, 16 head, 63 sec, 2503872
sectors
wd0(wdc0:0:0): using BIOS timings
wdc1 at isa0 port 0x170/8 irq 15
atapiscsi0 at wdc1 channel 0 drive 0
scsibus0 at atapiscsi0: 2 targets
wdc_atapi_intr: warning: reading only 14 of 18 bytes
wdc_atapi_intr: warning: reading only 0 of 58 bytes
wdc_atapi_intr: warning: reading only 14 of 18 bytes
cd0 at scsibus0 targ 0 lun 0: <TOSHIBA, CD-ROM XM-5302TA, 1095> SCSI0
5/cdrom removable
cd0(wdc1:0:0): using BIOS timings
pcppi0 at isa0 port 0x61
midi0 at pcppi0: <PC speaker>
sysbeep0 at pcppi0
lpt0 at isa0 port 0x378/4 irq 7
npx0 at isa0 port 0xf0/16: using exception 16
pccom0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo
pccom1 at isa0 port 0x2f8/8 irq 3: ns16550a, 16 byte fifo
fdc0 at isa0 port 0x3f0/6 irq 6 drq 2
fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec
biomask c040 netmask c840 ttymask d8c2
pctr: 586-class performance counters and user-level cycle counter
enabled
dkcsum: wd0 matched BIOS disk 80
root on wd0a
rootdev=0x0 rrootdev=0x300 rawdev=0x302
ts /root #

[prev in list] [next in list] [prev in thread] [next in thread]