| |||||||||||||
Please make us proud of ext4Please make us proud of ext4Posted Mar 17, 2009 17:49 UTC (Tue) by sbergman27 (guest, #10767)Parent article: Better than POSIX?
"allocate-on-commit" as the default, with a separate option to allow less reliable, but more performant behavior (perhaps alloc=benchmarks) would exhibit the traditional, and time-tested, extX philosophy that has worked so well. The ext3 devs were steadfast in insisting that data=ordered be the default for ext3, even though data=writeback is generally a performance win.
(Log in to post comments)
Please make us proud of ext4 Posted Mar 19, 2009 2:39 UTC (Thu) by mjr (guest, #6979) [Link]
Yes, make us proud, by actually using the filesystem's capabilities and not further discouraging the proper use of fsync (not overuse, such as firefox is/has been prone to...) by keeping it unnecessarily slow (which alloc-on-commit will do).
The kludges for alloc-on-replace that will apparently be enabled by default should deal with the real, problematic cases of loss of pre-existing data quite nicely, and I do think having those as defaults is a good, reasonable compromise. Those who want to be extra-paranoid with either nodelalloc or alloc-on-commit, well, sure, the option is/will be there, but I don't think these might I say somewhat obsolescent modes should be the default.
'course, distros shouldn't use ext4 per default yet anyway for conservativeness sake. Maybe starting next year ;]
|
|||||||||||||