Sun Solaris 10 ZFS offers blazing filesystem performance

The new Sun file system ZFS represents a previously unseen breakthrough in file system performance and management. All HELIOS UB based products and tools have successfully passed performance and reliability testing using the new Sun ZFS file system. Important tasks like working with many small files and re-indexing the HELIOS Desktop database work many times faster compared to the standard UFS file system. In addition, ZFS offers disk / file system / reliability features that are missing in other standard server operating systems. Our engineers are delighted and we feel that the latest Sun options like supporting AMD-CPUs, offering OpenSolaris on multiple platforms – and especially the new ZFS – bring unmatched performance benefits for customers.

For the following test run (3 million files tested), a directory tree is created, rebuilt and then removed. Click here to see performance test details.

 

Task	ZFS [min.]	UFS [min.]	ZFS:UFS performance
Create	3.75	93.75	25x faster
Rebuild HELIOS Desktop	25.30	116.75	4.6x faster
Remove	16.75	217.30	13x faster

 [min.] = minutes

• 128-bit file system(s); no practical size limit
  
• Transactional semantics (like journaling, disk is always consistent)
  
• Dynamic storage management
  
  • Allows growing file system size
    
  • Allows adding disks
    
  • Allows striping and mirroring
    
  • No dedicated file system space reservation required, e.g. two file systems can grow until ZFS pool is full
    
• File system snapshot support
  
  • Allows backup of the frozen system without service interrupt
    
  • Allows backup snapshots to allow easy access of older files, e.g. snapshot at 10 a.m., 12 a.m., 2 p.m., 4 p.m., etc.
    
  • Allows replicating a ZFS snapshot (e.g. to a remote disk / server)
    
  • Allows rolling back the file system to an older snapshot version 
    
• Snapshot cloning allows read-write access of a snapshot
  
• Disk block checksums will detect data errors
  
• Disk block compression features
  
• ZFS disk image compatibility between different platforms (SPARC / AMD)

• Sun SPARC / x86 (64-bit is highly recommended!)
  
• Solaris 10 (edition 11/06) or newer; earlier Solaris 10 ZFS versions will not work reliably
  
• 512 MB RAM (the more memory the better performance!)
  
• Any disk or RAID system will work
  
• HELIOS UB products (arch.: “sol4” and “solx86”)
  
• Attention: For using ZFS with database applications see TechInfo #106

HELIOS file servers use the AFP 3 file service (EtherShare UB) for Mac files and SMB/CIFS (PCShare UB) for Windows files. A Mac file is comprised of a data fork and a resource fork and Windows files can contain metadata and file streams. That means that each file in the file system can have several forks. Since ZFS can handle many small files way faster than UFS, HELIOS file services benefit from this performance boost.

In our example we will use a complete disk “c0t3d0” and later add another disk “c0t4d0” for the ZFS pool “testpool”. For this pool, a file system container “myprojects” will be created and a few properties set, which are inherited by the individual “project1” and “project2”.

 

1. Create a ZFS pool via:
   Format: zpool create poolname storage*
   zpool create testpool c0t3d0
   * storage can be diskname (c0t3d0), partitionname(c0td0s5), or filepath(/data/bigfile)
 
2. Create two file systems “project1” and “project2” on this pool
   Format: zfs create poolname / filesystemname
   First create a file system hierarchy, which acts as a container for the individual file systems that will be created later, and set required properties:
   # zfs create testpool/myprojects
   # zfs set mountpoint=/export/myprojects testpool/myprojects
   
   Then create your individual project file system(s):
   # zfs create testpool/myprojects/project1
   # zfs create testpool/myprojects/project2
   
   Both file systems “project1” and “project2” will automatically be mounted below “/export/myprojects”.
   
   Note: Where to define HELIOS volumes: at least one level below file system mount point!
   
   Example: For the “/export/myprojects/project1” file system the HELIOS mount point could be at “/export/myprojects/project1/project1_volume” but must NOT reside on the “project1” directory itself. This is required due to the way ZFS stores its snapshot information from a ZFS file system.
   
   Additional storage can be added via:
   Format: zpool add poolname storage
   
   Add a second disk “c0t4d0”:
   # zpool add testpool c0t4d0
    
3. Create a snapshot of the file system “project”:
   Format: zfs snapshot <poolname>/<filesystemname>@snapshotname
   # zfs snapshot testpool/myprojects/project1@snap
 
4. Create a clone read-write file system of the snapshot:
   Format: zfs clone <poolname>/<filesystemname>@snapshotname <poolname>/<filesystemname>
   # zfs clone testpool/myprojects/project1@snap testpool/myprojects/clone_p1
   The clone of “project1” is available at: “/export/myprojects/clone_p1”.
 
5. List all ZFS file systems:
   # zfs list
   
   

   NAME	USED	AVAIL	REFER	MOUNTPOINT
   test/…/project1	29.2M	53.8G	28.7M	/exp/…/project1
   test/…/project2	329.2M	53.8G	328.8M	/exp/…/project2
   test/…/project1@snap	505K	–	28.7M	–
   test/…/clone_p1	806K	53.8G	28.7M	/exp/…/clone_p1

 
6. Remove the clone, remove the snapshot:
   Format: zfs destroy <poolname>/<filesystemname>
   OR zfs destroy <poolname>/<filesystemname>@snapshotname
   # zfs destroy testpool/myprojects/clone_p1
   # zfs destroy testpool/myprojects/project1@snap
 
7. Backup a ZFS file system snapshot into a disk file:
   Format: zfs send <poolname>/<filesystemname>@<snapshotname> > filepath
   # zfs send testpool/myprojects/project1@snap > /backup/project1_snap.bkp
 
8. Remote HTTP based ZFS administration via:
   “https://hostname:6789/zfs”

Get I/O statistics

Tested applications

Tools 

Performance test details

Notes:

Additional ZFS links