Lenz Grimmer <grimmer@suse.de> writes:
But I must agree - ReiserFS is great. I have converted all my boxes (private and at work) to it and it runs flawlessly.
I would presume there is some decrease in speed. May I ask you, who are using it, if you found the decrease significantly noticeable? Did you find any other disadvantage? Better ask others who dived and tried, than discover the disadvantages ourselves, given the investment needed in reformatting and moving partitions around. Yet, to be more precise, most of the time is for reading documents, studying things, and taking preventive actions for being able to switch back in case we are not satisfied. How committed are the ReiserFS developers to stay back compatible with the current state, as the format evolves? If they are, this is a good incentive for trying it now. If not, this is an indication that we should wait. I just saw `ext3' mentioned, and have no more information. If someone happen to know, could the main features be sketched? Is `ext3' in the evolutionary line set by ReiserFS, or are these to be competing directions? Such information might help us to decide when (and where!) to switch. -- François Pinard http://www.iro.umontreal.ca/~pinard -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/
Reiser's web site has benchmarks that claim that it is a little faster the ext2. I personally did not notice any speed difference. Avi François Pinard wrote:
Lenz Grimmer <grimmer@suse.de> writes:
But I must agree - ReiserFS is great. I have converted all my boxes (private and at work) to it and it runs flawlessly.
I would presume there is some decrease in speed. May I ask you, who are using it, if you found the decrease significantly noticeable? Did you find any other disadvantage? Better ask others who dived and tried, than discover the disadvantages ourselves, given the investment needed in reformatting and moving partitions around. Yet, to be more precise, most of the time is for reading documents, studying things, and taking preventive actions for being able to switch back in case we are not satisfied.
-- Avi Schwartz Get a Life avi@CFFtechnologies.com Get Linux -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/
Hi, orry for the late reply. I just stumbled over this message in my inbox... On 28 Mar 2000, François Pinard wrote:
Lenz Grimmer <grimmer@suse.de> writes:
But I must agree - ReiserFS is great. I have converted all my boxes (private and at work) to it and it runs flawlessly.
I would presume there is some decrease in speed. May I ask you, who are using it, if you found the decrease significantly noticeable?
So far, I was not able to notice any significant loss of speed. Only if the machine having heavy disk activity, the mouse cursor in X is a bit sloppy. But all in all, there is no major difference (in any direction).
Did you find any other disadvantage? Better ask others who dived and tried, than discover the disadvantages ourselves, given the investment needed in reformatting and moving partitions around. Yet, to be more precise, most of the time is for reading documents, studying things, and taking preventive actions for being able to switch back in case we are not satisfied.
I have not found any other disadvantage yet, apart from needing to reformat the partitions :) We have converted most of our main development servers to reiserfs and are very satisfied so far.
How committed are the ReiserFS developers to stay back compatible with the current state, as the format evolves? If they are, this is a good incentive for trying it now. If not, this is an indication that we should wait.
There were some compatibility issues with the very first versions, but this has been resolved. We have urged the ReiserFS developers to stay backward compatible and they are committed to leave it that way.
I just saw `ext3' mentioned, and have no more information. If someone happen to know, could the main features be sketched? Is `ext3' in the evolutionary line set by ReiserFS, or are these to be competing directions?
Such information might help us to decide when (and where!) to switch.
IIRC, I described the main differences in another message. Basically the aim at the same target, but use different approaches. Bye, LenZ -- ------------------------------------------------------------------ Lenz Grimmer SuSE GmbH mailto:grimmer@suse.de Schanzaeckerstr. 10 http://www.suse.de/~grimmer 90443 Nuernberg, Germany -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/
I too like reiserfs. I've been using it in a development system for several months without a data failure. There is only one thing I noticed that makes it not perform like ext2. We have a data acquisition program running in the linux box. The data is stored to disk in fixed block sizes. At the same time another program is displaying the data on the monitor. This program checks the data file periodically for new blocks to display. When running on a reiserfs system, the update stops and I have to re-read the file to display the new blocks. In an ext2 system I don't see this behavior. It seems like the file's info is not updated quickly enough. On the other hand these data files can be large, several megabytes, and as I understand it, reiserfs is best suited for small files (although I don't know how small "small" is.) -- Rafael Herrera Laboratory for Computational Neuroscience University of Pittsburgh http://www.neuronet.pitt.edu/~raffo -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/
participants (4)
-
avi＠CFFtechnologies.com -
grimmer＠suse.de -
pinard＠iro.umontreal.ca -
raffo＠neuronet.pitt.edu