[solved] Kernel configuration for Heavy Calculations Servers

Question

Hello all,
I've been using Arch on half of the dual xeon workstations of the research lab i work to (these are the only ones under my supervision) and so far they have been continuously hailed for their greater performance compared to the other machines (same configuration) running fedora.
I did some minor tweaks on makepkg.conf (and using sugested compilation flags of the software i use) and on the bios to improve memory latency but i suppose i could squeeze even more juice from those boxes if i choose the right kernel options.
Finally my questions are:
- regarding the kernel, what options could i use to improve performance (i suppose using a home-pc kernel configuration is not the best option)?
- regarding the arch linux, what else can i do?
Usage:
- only to perform molecular modeling calculations (both molecular dynamics and quantum dynamics )
- data transfer using "scp"
Below are some important info about the hardware and config files:
-lspci
00:00.0 Host bridge: Intel Corporation 5000V Chipset Memory Controller Hub (rev b1)
00:02.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x8 Port 2-3 (rev b1)
00:03.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x4 Port 3 (rev b1)
00:08.0 System peripheral: Intel Corporation 5000 Series Chipset DMA Engine (rev b1)
00:10.0 Host bridge: Intel Corporation 5000 Series Chipset FSB Registers (rev b1)
00:10.1 Host bridge: Intel Corporation 5000 Series Chipset FSB Registers (rev b1)
00:10.2 Host bridge: Intel Corporation 5000 Series Chipset FSB Registers (rev b1)
00:11.0 Host bridge: Intel Corporation 5000 Series Chipset Reserved Registers (rev b1)
00:13.0 Host bridge: Intel Corporation 5000 Series Chipset Reserved Registers (rev b1)
00:15.0 Host bridge: Intel Corporation 5000 Series Chipset FBD Registers (rev b1)
00:16.0 Host bridge: Intel Corporation 5000 Series Chipset FBD Registers (rev b1)
00:1c.0 PCI bridge: Intel Corporation 631xESB/632xESB/3100 Chipset PCI Express Root Port 1 (rev 09)
00:1d.0 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB Controller #1 (rev 09)
00:1d.1 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB Controller #2 (rev 09)
00:1d.2 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB Controller #3 (rev 09)
00:1d.3 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB Controller #4 (rev 09)
00:1d.7 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset EHCI USB2 Controller (rev 09)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev d9)
00:1f.0 ISA bridge: Intel Corporation 631xESB/632xESB/3100 Chipset LPC Interface Controller (rev 09)
00:1f.1 IDE interface: Intel Corporation 631xESB/632xESB IDE Controller (rev 09)
00:1f.2 IDE interface: Intel Corporation 631xESB/632xESB/3100 Chipset SATA IDE Controller (rev 09)
00:1f.3 SMBus: Intel Corporation 631xESB/632xESB/3100 Chipset SMBus Controller (rev 09)
01:00.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Upstream Port (rev 01)
01:00.3 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express to PCI-X Bridge (rev 01)
02:00.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Downstream Port E1 (rev 01)
02:01.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Downstream Port E2 (rev 01)
02:02.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Downstream Port E3 (rev 01)
05:00.0 Ethernet controller: Intel Corporation 80003ES2LAN Gigabit Ethernet Controller (Copper) (rev 01)
05:00.1 Ethernet controller: Intel Corporation 80003ES2LAN Gigabit Ethernet Controller (Copper) (rev 01)
09:0c.0 VGA compatible controller: ATI Technologies Inc ES1000 (rev 02)
-cat /proc/cpuinfo - it's the same for the eight "processors"
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Xeon(R) CPU E5410 @ 2.33GHz
stepping : 6
cpu MHz : 2331.000
cache size : 6144 KB
physical id : 0
siblings : 4
core id : 0
cpu cores : 4
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 10
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 lahf_lm tpr_shadow vnmi flexpriority
bogomips : 4656.91
clflush size : 64
cache_alignment : 64
address sizes : 38 bits physical, 48 bits virtual
power management:
-/etc/rc.conf
MOD_AUTOLOAD="yes"
MODULES=(acpi-cpufreq )
USELVM="no"
DAEMONS=(syslog-ng hal sensorsd sshd network netfs crond)
-free -m
total used free shared buffers cached
Mem: 7968 3094 4874 0 75 2784
-/+ buffers/cache: 233 7734
Swap: 19077 0 19077
- modifications i made to the /etc/makepkg.conf
# ARCHITECTURE, COMPILE FLAGS
CARCH="x86_64"
CHOST="x86_64-unknown-linux-gnu"
#-- Exclusive: will only run on x86_64
# -march (or -mcpu) builds exclusively for an architecture
# -mtune optimizes for an architecture, but builds for whole processor family
CFLAGS="-march=native -mtune=native -03 -pipe"
CXXFLAGS="-march=native -mtune=native -03 -pipe"
#LDFLAGS=""
#-- Make Flags: change this for DistCC/SMP systems
MAKEFLAGS="-j9"
Thanks in advance
Last edited by Duca (2009-09-28 04:30:04)

Answer 1

graysky wrote:Your xfers are via scp... are the files large or numerous small files?  Have you the ability to use jumbo  frames on your network if the network is a GigLAN backbone?  If so, you'll have to test to see if a non-standard 4000 or 9000 mtu size improves xfers.
thanks for your rapid response and i googled about it, but unfortunatelly i dont have a gigLan in the lab but usually all file transfered exceed 5GB of size.
Would the scheduler have a significant impact on performance? The servers are used individually (not sharing load like in a bewoulf cluster)
The disk I/O operations are mostly about writes into the output files (a few number of bytes per write), would reiserfs be superior to ext3 ?
PS.: i'm feeling really stupid for not reading more about the kernel26-lts for it has all the major options i was thinking about.
Last edited by Duca (2009-09-27 18:26:34)