Oleg Drokin | 11c8779 | 2014-08-17 13:13:51 -0400 | [diff] [blame] | 1 | Lustre Parallel Filesystem Client |
| 2 | ================================= |
| 3 | |
| 4 | The Lustre file system is an open-source, parallel file system |
| 5 | that supports many requirements of leadership class HPC simulation |
| 6 | environments. |
| 7 | Born from from a research project at Carnegie Mellon University, |
| 8 | the Lustre file system is a widely-used option in HPC. |
| 9 | The Lustre file system provides a POSIX compliant file system interface, |
| 10 | can scale to thousands of clients, petabytes of storage and |
| 11 | hundreds of gigabytes per second of I/O bandwidth. |
| 12 | |
| 13 | Unlike shared disk storage cluster filesystems (e.g. OCFS2, GFS, GPFS), |
| 14 | Lustre has independent Metadata and Data servers that clients can access |
| 15 | in parallel to maximize performance. |
| 16 | |
Andreas Dilger | d98229f | 2015-08-24 11:11:49 -0400 | [diff] [blame] | 17 | In order to use Lustre client you will need to download the "lustre-client" |
| 18 | package that contains the userspace tools from http://lustre.org/download/ |
Oleg Drokin | 11c8779 | 2014-08-17 13:13:51 -0400 | [diff] [blame] | 19 | |
| 20 | You will need to install and configure your Lustre servers separately. |
| 21 | |
| 22 | Mount Syntax |
| 23 | ============ |
| 24 | After you installed the lustre-client tools including mount.lustre binary |
| 25 | you can mount your Lustre filesystem with: |
| 26 | |
| 27 | mount -t lustre mgs:/fsname mnt |
| 28 | |
| 29 | where mgs is the host name or ip address of your Lustre MGS(management service) |
| 30 | fsname is the name of the filesystem you would like to mount. |
| 31 | |
| 32 | |
| 33 | Mount Options |
| 34 | ============= |
| 35 | |
| 36 | noflock |
| 37 | Disable posix file locking (Applications trying to use |
| 38 | the functionality will get ENOSYS) |
| 39 | |
| 40 | localflock |
| 41 | Enable local flock support, using only client-local flock |
| 42 | (faster, for applications that require flock but do not run |
| 43 | on multiple nodes). |
| 44 | |
| 45 | flock |
| 46 | Enable cluster-global posix file locking coherent across all |
| 47 | client nodes. |
| 48 | |
| 49 | user_xattr, nouser_xattr |
| 50 | Support "user." extended attributes (or not) |
| 51 | |
| 52 | user_fid2path, nouser_fid2path |
| 53 | Enable FID to path translation by regular users (or not) |
| 54 | |
| 55 | checksum, nochecksum |
| 56 | Verify data consistency on the wire and in memory as it passes |
| 57 | between the layers (or not). |
| 58 | |
| 59 | lruresize, nolruresize |
| 60 | Allow lock LRU to be controlled by memory pressure on the server |
| 61 | (or only 100 (default, controlled by lru_size proc parameter) locks |
| 62 | per CPU per server on this client). |
| 63 | |
| 64 | lazystatfs, nolazystatfs |
| 65 | Do not block in statfs() if some of the servers are down. |
| 66 | |
| 67 | 32bitapi |
| 68 | Shrink inode numbers to fit into 32 bits. This is necessary |
| 69 | if you plan to reexport Lustre filesystem from this client via |
| 70 | NFSv4. |
| 71 | |
| 72 | verbose, noverbose |
| 73 | Enable mount/umount console messages (or not) |
| 74 | |
| 75 | More Information |
| 76 | ================ |
Andreas Dilger | d98229f | 2015-08-24 11:11:49 -0400 | [diff] [blame] | 77 | You can get more information at the Lustre website: http://wiki.lustre.org/ |
Oleg Drokin | 11c8779 | 2014-08-17 13:13:51 -0400 | [diff] [blame] | 78 | |
Andreas Dilger | d98229f | 2015-08-24 11:11:49 -0400 | [diff] [blame] | 79 | Source for the userspace tools and out-of-tree client and server code |
| 80 | is available at: http://git.hpdd.intel.com/fs/lustre-release.git |
Oleg Drokin | 11c8779 | 2014-08-17 13:13:51 -0400 | [diff] [blame] | 81 | |
| 82 | Latest binary packages: |
Andreas Dilger | d98229f | 2015-08-24 11:11:49 -0400 | [diff] [blame] | 83 | http://lustre.org/download/ |