> On Thu, Aug 06, 1998 at 03:56:16PM -0500, Shawn Leas wrote:
>
> > One with millions of inodes even with btree will be slow. I've
> > benchmarked reiserfs, have you???
>
> Can you supply more details?
>
> Unless you btree is hosed, searching (say) 10 million records for a
> key should be pretty fast...
reiserfs was about 30 times faster than ext2 doing simplistic benchmarks
that really aren't fair.... This means, reiser was doing what it did
best, against what ext2 does worst...
Here is an e-mail I sent to the reiserfs list a while back, and they have
made improvements since then I might add... Pretty damn quick FS!
The tests consisted simply of doing some operation on
a ton of directories. In the root of the test dir, there
were dirs a,b,c,d,e, and f. Each of these had dirs named
0-9, and these all had about 881 subdirs. So, there were
52867 dirs total. Why all dirs?? Guess I wanted to go home.
BTW, the ext2 was mounted noatime, and nodiratime, as
the mounts info below will show.
The first test:
find | wc -l
The second test:
du -ks ?
The third test:
rm -rf ?
In short, this kind of operation seems to be where
either I've tuned my HD out, or the reiserfs kicks
arse.
-------------------------------------------------------------
Time Time ext2 /
TEST ext2 reiser reiser
-------------------------------------------------------------
find 390.02 11.599 33.6253125269
du -ks ? 467.99 11.578 40.4206253238
rm -rf ? 573.46 27.970 20.5026814444
-------------------------------------------------------------
AVG = 31.5162064317
I have included some relevant system information. The
hdparm I have setup probably had an adverse affect on
performance for the ext2, because of the large fs-
readahead. I didn't want to re-tune for the test
after I realized this. OH well, we get the point...
(to follow are mounts, dmesg, and hdparm output
from when I tuned my hard drive a little)
**********************************************
MOUNTS
**********************************************
/dev/hda8 on / type ext2 (rw,noatime,nodiratime)
none on /proc type proc (rw)
/dev/hda1 on /dos/c type vfat (rw)
/dev/hda5 on /dos/d type vfat (rw)
/dev/hda6 on /dos/e type reiserfs (rw)
**********************************************
DMESG
**********************************************
Linux version 2.1.106 (root@berzerk.mn26.integris-ds.com) (gcc version pgcc-2.90.29 980515 (egcs-1.0.3 release)) #2 Wed Jun 17 01:23:35 CDT 1998
Console: 16 point font, 400 scans
Console: colour VGA+ 80x25, 1 virtual console (max 63)
Calibrating delay loop... 232.65 BogoMIPS
Memory: 63296k/65536k available (856k kernel code, 392k reserved, 956k data, 36k init)
CPU: Intel Pentium II (Klamath) stepping 04
Checking 386/387 coupling... Ok, fpu using exception 16 error reporting.
Checking 'hlt' instruction... Ok.
mtrr: v1.21 (19980521) Richard Gooch (rgooch@atnf.csiro.au)
POSIX conformance testing by UNIFIX
Swansea University Computer Society NET3.039 for Linux 2.1
Swansea University Computer Society TCP/IP for NET3.037
IP Protocols: ICMP, UDP, TCP
Initializing RT netlink socket
PCI: PCI BIOS revision 2.10 entry at 0xfd9cc
PCI: Using configuration type 1
PCI: Probing PCI hardware
Starting kswapd v 1.5
Serial driver version 4.26 with no serial options enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
APM BIOS version 1.2 Flags 0x03 (Driver version 1.4)
Entry f000:4c88 cseg16 f000 dseg 40 cseg len ffff, cseg16 len ffff, dseg len ffff
Connection version 1.2
AC on line, battery status unknown, battery life unknown
battery flag 0x80, battery life unknown
Real Time Clock Driver v1.09
PIIX4: IDE controller on PCI bus 00 dev 39
PIIX4: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xfc90-0xfc97, BIOS settings: hda:DMA, hdb:pio
ide1: BM-DMA at 0xfc98-0xfc9f, BIOS settings: hdc:DMA, hdd:pio
hda: Maxtor 86480D6, ATA DISK drive
hdc: TOSHIBA CD-ROM XM-6202B, ATAPI CDROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: Maxtor 86480D6, 6149MB w/256kB Cache, CHS=784/255/63, UDMA
Floppy drive(s): fd0 is 1.44M
FDC 0 is a National Semiconductor PC87306
md driver 0.36.6 MAX_MD_DEV=4, MAX_REAL=8
Partition check:
hda: hda1 hda2 < hda5 hda6 hda7 hda8 >
VFS: Mounted root (ext2 filesystem) readonly.
Freeing unused kernel memory: 36k freed
Adding Swap: 72256k swap-space (priority -1)
NET3: Unix domain sockets 0.16 for Linux NET3.038.
3c59x.c:v0.47H 12/4/97 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
eth0: 3Com 3c905 Boomerang 100baseTx at 0xfcc0, 00:10:4b:1f:6c:37, IRQ 10
8K word-wide RAM 3:5 Rx:Tx split, autoselect/MII interface.
eth0: MII transceiver found at address 24.
Enabling bus-master transmits and whole-frame receives.
PS/2 auxiliary pointing device detected -- driver installed.
reiserfs_read_super: use 1024 as initial blocksize
**********************************************
HDPARM
**********************************************
/dev/hda:
setting fs readahead to 32
setting 32-bit I/O support flag to 1
setting multcount to 16
setting unmaskirq to 1 (on)
setting keep_settings to 1 (on)
setting drive keep features to 1 (on)
setting drive write-caching to 1 (on)
multcount = 16 (on)
I/O support = 1 (32-bit)
unmaskirq = 1 (on)
keepsettings = 1 (on)
readahead = 32 (on)
Model=Maxtor 86480D6, FwRev=NAVX171F, SerialNo=L60HS4CA
Config={ Fixed }
RawCHS=13328/15/63, TrkSize=0, SectSize=0, ECCbytes=20
BuffType=3(DualPortCache), BuffSize=256kB, MaxMultSect=16, MultSect=16
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
CurCHS=13328/15/63, CurSects=12594960, LBA=yes, LBAsects=12594960
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
-Shawn
<=========== America Held Hostage ===========>
Day 2025 for the poor and the middle class.
Day 2044 for the rich and the dead.
897 days remaining in the Raw Deal.
<============================================>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html