Databases
StackOverflow2013
Postsdbo
POZFS on Linux, getting slow write?
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POZFS on Linux, getting slow write?
    primarykey
    data
    text
    <p>Decided to give ZFS on linux v28 a go (native port ubuntu) on my 5x 2TB WD Green advanced format drives. Created the pool using "zpool create -o ashift=12 raidz1 "</p> <p>zpool status doesnt show anything unsual.</p> <p>I did a dd if=/dev/zero into the mounted pool, and i can never reach past 20M/s write. I attempted to rsync a couple hundred gigs of files but even then, 'zpool iostat' gives me a maximum of 20M write. Cpu usage isnt very high and my 8GB of ram is 90% utilised – which I believe is normal.</p> <p>Reading performance seems optimal to me.</p> <p>I did play around with zfs_vdev_max/min_pending. As I have AHCI enabled I attempted setting these values to 1 but that reduced my writes down to 10M. Bringing it up to min/max of 4/8 reverted it back to 20M write speed.</p> <p>Im doing a scrub now, and thats going at 170M/s.</p> <p>Im thinking there must be something I missed ? Or is this normal ?</p> <p>Attached are my settings. Ignore the sparse file, saving it to replace with a disk later.</p> <p>zdb:</p> <pre><code>myData: version: 28 name: 'myData' state: 0 txg: 12 pool_guid: 14947267682211456191 hostname: 'microserver' vdev_children: 1 vdev_tree: type: 'root' id: 0 guid: 14947267682211456191 create_txg: 4 children[0]: type: 'raidz' id: 0 guid: 361537219350560701 nparity: 1 metaslab_array: 31 metaslab_shift: 36 ashift: 12 asize: 10001923440640 is_log: 0 create_txg: 4 children[0]: type: 'file' id: 0 guid: 18296057043113196254 path: '/tmp/sparse2' DTL: 35 create_txg: 4 offline: 1 children[1]: type: 'disk' id: 1 guid: 13192250717230911873 path: '/dev/disk/by-id/wwn-0x50014ee2062a07cd-part1' whole_disk: 1 create_txg: 4 children[2]: type: 'disk' id: 2 guid: 7673445363652446830 path: '/dev/disk/by-id/wwn-0x50014ee25bd8fbcc-part1' whole_disk: 1 create_txg: 4 children[3]: type: 'disk' id: 3 guid: 1997560602751946723 path: '/dev/disk/by-id/wwn-0x50014ee25b1edbc8-part1' whole_disk: 1 create_txg: 4 children[4]: type: 'disk' id: 4 guid: 16890030280879643154 path: '/dev/disk/by-id/wwn-0x50014ee25b7f2562-part1' whole_disk: 1 create_txg: 4 zfs get all myData: NAME PROPERTY VALUE SOURCE myData type filesystem - myData creation Tue Apr 24 8:15 2012 - myData used 2.05T - myData available 5.07T - myData referenced 2.05T - myData compressratio 1.00x - myData mounted yes - myData quota none default myData reservation none default myData recordsize 128K default myData mountpoint /myData default myData sharenfs off default myData checksum on default myData compression off default myData atime on default myData devices on default myData exec on default myData setuid on default myData readonly off default myData zoned off default myData snapdir hidden default myData aclinherit restricted default myData canmount on default myData xattr on default myData copies 1 default myData version 5 - myData utf8only off - myData normalization none - myData casesensitivity sensitive - myData vscan off default myData nbmand off default myData sharesmb off default myData refquota none default myData refreservation none default myData primarycache all default myData secondarycache all default myData usedbysnapshots 0 - myData usedbydataset 2.05T - myData usedbychildren 9.68M - myData usedbyrefreservation 0 - myData logbias latency default myData dedup off default myData mlslabel none default myData sync standard default myData refcompressratio 1.00x - zpool get all myData: NAME PROPERTY VALUE SOURCE myData size 9.06T - myData capacity 28% - myData altroot - default myData health DEGRADED - myData guid 14947267682211456191 default myData version 28 default myData bootfs - default myData delegation on default myData autoreplace off default myData cachefile - default myData failmode wait default myData listsnapshots off default myData autoexpand off default myData dedupditto 0 default myData dedupratio 1.00x - myData free 6.49T - myData allocated 2.57T - myData readonly off - myData ashift 12 local </code></pre>
    singulars
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PTQuestion
    1. This table or related slice is empty.
    1. USKambo_Rambo
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. POZFS on Linux, getting slow write?
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POZFS on Linux, getting slow write?
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. POZFS on Linux, getting slow write?
      1. USsenorsmile
      1. VTFavorite
    1. COShould be moved to unix.stackoverflow.com.
      singulars
      1. POZFS on Linux, getting slow write?
      1. USWarren Young
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload