sarah@clusterfs.com
2006-Dec-15 21:55 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Finish kern, dd and dbench tests on DDN with sector size=512. Encouter Kernel BUG when testing dbench256(EXT4) and dbench512(EXT4). Report(report-DDN-512) and error log(dbench-errorlog-512) please see the attached.
sarah@clusterfs.com
2006-Dec-15 21:57 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Created an attachment (id=9155) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9155&action=view) report for DDN with sector size=512
sarah@clusterfs.com
2006-Dec-15 21:58 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Created an attachment (id=9156) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9156&action=view) error log when testing dbench256/512 with EXT4
sarah@clusterfs.com
2006-Dec-21 19:45 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Created an attachment (id=9205) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9205&action=view) report for DDN with sector size=4096
sarah@clusterfs.com
2006-Dec-28 20:07 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Created an attachment (id=9239) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9239&action=view) report for thumper
sarah@clusterfs.com
2007-Jan-11 03:09 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Created an attachment (id=9316) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9316&action=view) test result with dbench v2.1
alex@clusterfs.com
2007-Jan-12 07:31 UTC
[Lustre-devel] [Bug 10816] reservation based allocator
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 Created an attachment (id=9324) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9324&action=view) test plan (DRAFT) we still need to describe few more tests with more realistic load (dbench, for example)
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=10816 (In reply to comment #19)> test plan (DRAFT)Suggestions: o Very large file (8 TB?) corner case o Comparison of results vs. current allocator o Use various different disk back-ends o Allocator testing as part of a Linux file system - Performance - Fragmentation