Re: Performance problems. Your opinion please !

r31dmaeu@xxxxxxxx wrote:
The environment, in short:

2 CPU's, hyper-threading active . Extract from the booting log:

CPU: Trace cache: 12K uops, L1 D cache: 8K
CPU: L2 cache: 512K
CPU: L3 cache: 4096K
CPU1: Initial APIC ID: 5, Physical Processor ID: 2
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#1.
CPU1: Intel P4/Xeon Extended MCE MSRs (12) available
CPU1: Thermal monitoring enabled
CPU1: Intel(R) Xeon(TM) MP CPU 3.00GHz stepping 06
Calibrating delay using timer specific routine.. 5987.46 BogoMIPS
CPU: Trace cache: 12K uops, L1 D cache: 8K
CPU: L2 cache: 512K
CPU: L3 cache: 4096K

Further:
Dual SCSI controller, 2 disk drives.
2 NFS mounted file systems.

The problem: The load average reaches values up to 25 or more.

At the time where problems occurs, the following approx. values are
observed, in a typical case, according to the "sar" log recording:

%CPU : user 16 %, system 8 %, iowait 60 % <---<<<
Total number of user sockets: 2000
Process list size: 500
Processes created: 5/s
Context switching: 2200 /s
Run queue size: 4
Number of unused cache entries in the directory cache: 5000
Number of used file handles: 15500

Further, the output of "cat /proc/sys/fs/file-nr" gives
"15645 0 371318".

Any idea where the problem is located ?

Sure: the system is waiting for IO.

Which system parameter should be changed ?

Thanks a lot !



--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ PGP-Key: 9A2FC99A Registered Machine 241939.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 07:30:01 up 25 days, 10:20, 4 users, load average: 4.12, 4.21, 4.13
.

Relevant Pages

  • IPv6 oops on ifup in latest BK
    ... Using ACPI for SMP configuration information ... Initializing CPU#0 ... CPU: Trace cache: 12K uops, ...
    (Linux-Kernel)
  • SCSI CDROM issue in kernels >= 2.6.14-rc3
    ... CPU: Trace cache: 12K uops, ... MEM window: disabled. ... SCSI device sda: 17928698 512-byte hdwr sectors ...
    (Linux-Kernel)
  • mptscsih: ioc1: attempting task abort! (sc=d6e8a980)
    ... CPU 2: Machine Check Exception: 0000000000000004 ... OEM ID: INTEL Product ID: Bridge CRB APIC at: 0xFEE00000 ... CPU: Trace cache: 12K uops, ... SCSI device sda: 287132440 512-byte hdwr sectors ...
    (Linux-Kernel)
  • Re: 2.6.16-rc6-mm2
    ... CPU: Trace cache: 12K uops, ... Calibrating delay using timer specific routine.. ... # ACPI Support ...
    (Linux-Kernel)
  • 2.6.16-rc5 huge memory detection regression
    ... I just tested 2.6.16-rc5 kernel on MSI 9136 dual Xeon server motherboard with 16 GB of memory and the kernel detects only 8 GB of RAM instead. ... CPU: Trace cache: 12K uops, ... SCSI device sda: 390721968 512-byte hdwr sectors ...
    (Linux-Kernel)