Skip to content

Test Plan for 4096 disks

Test Plan

  1. Objective
    1. Demonstrate that a Linux system can address, probe and do some amount of IO on a disk subsystem with 4096 disk
    2. Test this feature in conjunction with other features, including udev, filesystems, LVM, EVMS
  2. Outputs
    1. Once code is mature and equipment is available, integration test with multipath IO, Hotplug CPU, Hotplug memory, Hotplug IO Subsystem.
    2. When equipment is available on new architectures or IO sub-systems, reuse the scripts to quickly repeat the tests
    3. Reuse the tests to do regression tests at yet to be defined points in Linux 2.6 development
  3. Phases
    1. Phase One - ISCSI testing at OSDL for script development. - Mostly DONE - markh patches to iscsi list?
    2. Phase Two - Physical device testing at IBM. Test with udev filesystems ext3 and reiser, LVM and EVMS. - Tested 256 physical devices with ext3
    3. Phase Three - Post request for machine time on on other architectures and io subsystems and utilize as they become available. Integration with the other products mentioned above as equipment and/or code becomes available.
  4. Gating Factors
    1. Phase One -
      • working iSCSI targe code ( unh|| Netapp ) DONE
    2. Phase two -
      • configured system w/ 4096 physical LUNs
    3. Phase three
      1. donations of other equipment
      2. Maturity of Multipath I/O, Hotplug features.
      3. Test suites for MPIO and Hotplug.

Actions

  1. Phase one
    1. Get unh-iscsi target code running on OSDL system - DONE
    2. Create scripts to parse dmesg and count LUNs - DONE
    3. Create scripts to configure large number of LUNs on target - DONE
    4. Create/understand process of discovering LUNs on client - DONE
    5. Create scripts to install and configure large number of ext3 or reiser filesystems on target. - DONE
    6. Create scripts to install and configure udev on system. ( NoT NEEDED )
    7. Create scripts to install and configure LVM or EVMS on system.
    8. Create IO test harness with iozone and/or disktest ( DONE )
    9. Run demonstration at OSDL, discover/fix code issues ( DONE )
    10. Repeat until OSDL rig capacity exhausted.
  2. Phase Two
    1. Document IBM setup, ensure it can be torn down and rebuilt ( DONE )
    2. Run iozone/disktest on small number of spindles to establish baseline IO performance ( DONE )
    3. Configure full 4k spindles. ( DONE @ 1024 spindles )
    4. Re-run IO tests to verify raw performance. ( DONE @ 1024 spindles )
    5. Install and configure udev on 4k spindles. Run raw io tests against udev devices. ( DONE 1024 spindles )
    6. Install and configure ext3 file systems. Run buffered and unbuffered io tests against the file system. Same for reiser. ( DONE for ext3 @ 1024 spindles )
    7. Install and configure LVM. Run raw io tests against the LVM devices. Same for EVMS.