Difference between revisions of "Storage-Benches"

Un article de lcgwiki.
Jump to: navigation, search
(HP Systems)
(DELL Systems)
Ligne 6: Ligne 6:
  
 
== DELL Systems ==
 
== DELL Systems ==
 +
=== R510 + PowerVault MD1200 ===
 +
*RAID avec stripe size de 1m et règle « lecture anticipée adaptative »
 +
 +
*FS xfs avec alignement des partitions et montage avec l'option "noatime"
 +
 +
Exemple:
 +
#parted  /dev/sdb mklabel gpt
 +
#parted /dev/sdb mkpart primary xfs 1m 50%
 +
#parted /dev/sdb mkpart primary xfs 50% 100%
 +
#######
 +
mkfs.xfs -d su=1m,sw=10 /dev/sdb1 -L R510_sdb1
 +
mkfs.xfs -d su=1m,sw=10 /dev/sdb2 -L R510_sdb2
 +
#######
 +
# cat /etc/fstab | grep noatime
 +
LABEL=R510_sdb1          /fs1                    xfs    defaults,noatime        0 0
 +
LABEL=R510_sdb2          /fs2                    xfs    defaults,noatime        0 0
 +
 +
 +
*kernel:
 +
**Utilisation du Read ahead system
 +
**modif du scheduleur par défaut et du nombre de requete en queue
 +
 +
# tail -6 /etc/rc.d/rc.local
 +
blockdev --setra 16384 /dev/sdb
 +
blockdev --setra 16384 /dev/sdc
 +
echo 512 > /sys/block/sdb/queue/nr_requests
 +
echo 512 > /sys/block/sdc/queue/nr_requests
 +
echo deadline > /sys/block/sdb/queue/scheduler
 +
echo deadline > /sys/block/sdc/queue/scheduler
  
 
== HP Systems ==
 
== HP Systems ==

Version du 16:26, 8 juin 2011

Storage benchmarking and tuning depends upon several parameters like file systems, kernel parameters, RAID configuration, network parameters, etc. This page aims to provide benchmarking tools and procedures for LCG Storage Element, as well as some tuning hints for particular hardware solutions.

Benchmarking

Tuning

DELL Systems

R510 + PowerVault MD1200

  • RAID avec stripe size de 1m et règle « lecture anticipée adaptative »
  • FS xfs avec alignement des partitions et montage avec l'option "noatime"

Exemple:

#parted  /dev/sdb mklabel gpt
#parted /dev/sdb mkpart primary xfs 1m 50%
#parted /dev/sdb mkpart primary xfs 50% 100%
#######
mkfs.xfs -d su=1m,sw=10 /dev/sdb1 -L R510_sdb1
mkfs.xfs -d su=1m,sw=10 /dev/sdb2 -L R510_sdb2
#######
# cat /etc/fstab | grep noatime
LABEL=R510_sdb1          /fs1                    xfs    defaults,noatime        0 0
LABEL=R510_sdb2          /fs2                    xfs    defaults,noatime        0 0


  • kernel:
    • Utilisation du Read ahead system
    • modif du scheduleur par défaut et du nombre de requete en queue
# tail -6 /etc/rc.d/rc.local
blockdev --setra 16384 /dev/sdb
blockdev --setra 16384 /dev/sdc
echo 512 > /sys/block/sdb/queue/nr_requests
echo 512 > /sys/block/sdc/queue/nr_requests
echo deadline > /sys/block/sdb/queue/scheduler
echo deadline > /sys/block/sdc/queue/scheduler

HP Systems

MDS 600 based system

In order to enhance the performance of the MDS 600 based systems, the following tunings have been applied:

  • Upgrading the hard disk firmware (HPD3). All hard disks should have the same firmware version.
  • Changing the power management profile (max cpu power).
  • Using the following system parameters for each disk:
  echo "cfq" > /sys/block/cciss\!${disk}/queue/scheduler
  echo 256 > /sys/block/cciss\!${disk}/queue/nr_requests
  echo 4096 >  /sys/block/cciss\!${disk}/queue/read_ahead_kb
  • Using XFS filesystem aligned with the RAID strip
  • Using RAID units composed of 5 disks wisely selected on different columns.