Re: Why Debian3.1 is slower so much than FreeBSD6.1 in our test?
hi, Dave,
Just wondering whether you'd tried changing the filesystem on the Debian
system, or tweaking the parameters for it? This is one obvious
difference between your two setups, given that FreeBSD is using ufs.
the sginfo output is followed:
mysql-test-1:/home/huangjy# sginfo -A /dev/sda
INQUIRY reponse (cmd: 0x12)
---------------------------
Device Type 0
Vendor: FUJITSU
Product: MAP3367NC
Revision level: 5605
Serial Number 'UPL3P2B01LEY'
Read-Write Error Recovery mode page (0x1)
-----------------------------------------
AWRE 1
ARRE 1
TB 0
RC 0
EER 0
PER 0
DTE 0
DCR 0
Read Retry Count 63
Correction Span 233
Head Offset Count 0
Data Strobe Offset Count 0
Write Retry Count 63
Recovery Time Limit 30000
Disconnect-Reconnect mode page (0x2)
------------------------------------
Buffer full ratio 0
Buffer empty ratio 0
Bus Inactivity Limit 1
Disconnect Time Limit 0
Connect Time Limit 0
Maximum Burst Size 0
EMDP 0
Fair Arbitration (fcp:faa,fab,fac) 0
DIMM 0
DTDC 0
First Burst Size 0
Format Device mode page (0x3)
-----------------------------
Tracks per Zone 28
Alternate sectors per zone 84
Alternate tracks per zone 0
Alternate tracks per lu 36
Sectors per track 936
Data bytes per physical sector 512
Interleave 1
Track skew factor 326
Cylinder skew factor 102
Supports Soft Sectoring 0
Supports Hard Sectoring 1
Removable Medium 0
Surface 0
Rigid Disk Geometry mode page (0x4)
-----------------------------------
Number of cylinders 48122
Number of heads 2
Starting cyl. write precomp 0
Starting cyl. reduced current 0
Device step rate 0
Landing Zone Cylinder 0
RPL 0
Rotational Offset 0
Rotational Rate 10025
Verify Error Recovery mode page (0x7)
-------------------------------------
EER 0
PER 0
DTE 0
DCR 0
Verify Retry Count 63
Verify Correction Span (bits) 233
Verify Recovery Time Limit (ms) 30000
Caching mode page (0x8)
-----------------------
Initiator Control 0
ABPF 0
CAP 0
DISC 1
SIZE 0
Write Cache Enabled 1
MF 0
Read Cache Disabled 0
Demand Read Retention Priority 0
Demand Write Retention Priority 0
Disable Pre-fetch Transfer Length 65535
Minimum Pre-fetch 0
Maximum Pre-fetch 0
Maximum Pre-fetch Ceiling 65535
FSW 1
LBCSS 0
DRA 0
Number of Cache Segments 8
Cache Segment size 0
Non-Cache Segment size 0
Control mode page (0xa)
-----------------------
TST 0
TMF_ONLY 0
D_SENSE 0
GLTSD 0
RLEC 0
Queue Algorithm Modifier 0
QErr 0
DQue [obsolete] 0
TAS 0
RAC 0
UA_INTLCK_CTRL 0
SWP 0
RAERP [obs.] 0
UAAERP [obs.] 0
EAERP [obs.] 0
ATO 0
AUTOLOAD MODE 0
Ready AER Holdoff Period [obs.] 0
Busy Timeout Period 0
Extended self-test completion time 1107
Notch and Partition mode page (0xc)
-----------------------------------
Notched Drive 0
Logical or Physical Notch 0
Max # of notches 18
Active Notch 0
Starting Boundary 0x0
Ending Boundary 0xbbf901
Pages Notched 00000000 00000008
SPI-4 port control mode page (0x19)
-----------------------------------
Protocol identifier 1
Synchronous transfer time-out 0
SPI-4 Margin control mode subpage (0x19,0x1)
--------------------------------------------
Protocol identifier 1
Driver Strength 0
Driver Asymmetry 0
Driver Precompensation 0
Driver Slew rate 0
SPI-4 Negotiated Settings mode subpage (0x19,0x3)
--------------------------------------------
Transfer period 9
REQ/ACK offset 127
Transfer width exponent 1
Protocol option bits 2
Transciever mode 2
Sent PCOMP_EN 0
Received PCOMP_EN 0
SPI-4 Report transfer capabilities mode subpage (0x19,0x4)
--------------------------------------------
Mimimum transfer period factor 8
Maximum REQ/ACK offset 127
Maximum transfer width exponent 1
Protocol option bits supported 199
Informational Exceptions mode page (0x1c)
-----------------------------------------
PERF 0
EBF 0
EWASC 0
DEXCPT 0
TEST 0
LOGERR 0
MRIE 6
Interval Timer 6000
Report Count 0
mode page: 0x21
---------------
0x02 0x0f
0x03 0x00
ext3 flushes to disk every 5 seconds by default (see 'commit' interval
in 'man mount') which is a safe, conservative setting but might hit
performance.
The data is critical to us, so we must ensure every transaction is not lost.
Regards,
Leo Huang
2006/6/29, Dave Ewart <davee@sungate.co.uk>:
On Thursday, 29.06.2006 at 10:50 +0800, Leo Huang wrote:
> Any comment is welcome!
Just wondering whether you'd tried changing the filesystem on the Debian
system, or tweaking the parameters for it? This is one obvious
difference between your two setups, given that FreeBSD is using ufs.
ext3 flushes to disk every 5 seconds by default (see 'commit' interval
in 'man mount') which is a safe, conservative setting but might hit
performance.
Dave
--
Please don't CC me on list messages!
...
Dave Ewart - davee@sungate.co.uk - jabber: davee@jabber.org
All email from me is now digitally signed, key from http://www.sungate.co.uk/
Fingerprint: AEC5 9360 0A35 7F66 66E9 82E4 9E10 6769 CD28 DA92
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
iD8DBQFEo5+5nhBnac0o2pIRAszMAJ0bzPT1wCyO/WWjqheoBCRgj11iZwCgsBNz
0XhYk7bnfsRtZeAJlnQLN98=
=Izl2
-----END PGP SIGNATURE-----
Reply to: