From Slashdot: “Hans Reiser has benchmarked Reiser4 against ext3 and Reiserfs 3. Reiser4 turns out to be way faster than V3, and for ext3, why don’t you check out the results yourself ? Han’s Reiser states, “these benchmarks mean to me that our performance is now good enough to ship V4 to users”, and he will be probably sending in a patch within the next couple of weeks to be included in the 2.6/2.5 kernel.” Benchmarks against XFS would be most interesting too.
Reiser4 Benchmarks Available
2003-07-26 General Development 30 Comments
Reiser4 all the way.
im not a windows fan but just wondering has anyone ever compared reiser with any of these fs ??
i have been using the reiser file system for the last
few years and its great. the man is a genius. oh yes.
Can’t be compared, Windows can’t run on a Reiser4 partition, and Linux doen’t run well on FAT/NTFS partitions.
the performance of the same operations could be compared on the same machine…which no optimizations or max optimizations to each filesystem.
Still, could be a problem to choose the distro, kernel, KDE or GNOME, there are so many factors, but I guess it could be done.
So he thinks that great speed is why it should be shipped to users ? What about stability, data integrity and so on ? I will not take any chances with my data. ext2/3 for me for quite another while..
He think he have reached a good mix of usability and speed increase enough to make it a nice upgrade to V3, meaning it’s usable enough for public daredevils to beta test unlike before when we had to fetch it from his bk tree and do everything the hard way (to discourage too many to ruin partitions with it).
With other words, it can’t reach the stability you are looking for without releasing like this.
If I already have a resierfs partition(v3 i suppose) how would I go about updating it? would i have to reformat??
I suppose they’ll release tools to convert v3.6 partitions (that’s the version you probably use) to v4.
One thing that I like in linux is the possibility of choice between many filesystems. Because of linux modularity, any company or free project can create another inovative filesystem.
Windows NT is dependent only of NTFS filesystem and you cannot port it to anything but M$ OSes.
Some punk nicknamed “hansreiser” had this to say at Slashdot. Who knows if we should trust him…:
No, V4 is not backward compatible with V3. V3 and V4 are kept as separate codebases so that the new V4 features don’t destabilize V3. We are very serious about avoiding adding new features to V3, so that it can become a zero defect product.
However, there is a tool called convertfs (as well as tar) which can convert V3 to V4. It can also convert ext2 to V3 or V4 or V3 or V4 to ext2. It is pretty clever (and written by someone outside our team), in that it creates a loop back mounted target filesystem inside a file inside the source filesystem, copies everything from the source to the target, and then reshuffles the blocks of the file so that they are at the offsets on the device that they were at within the file.
Hope it’s going to make 2.6.0 though…
I’m not sure I’m going to use it…
But I just hate patching stuff…
So let’s pray it will get included in the vanilla linux kernel, if only to be marked “EXPERIMENTAL”
I use it if the distro allows it without further config. Otherwise I use ext3.
I prefer JFS or XFS when available but, in my experience, some distros tune the kernel (patch) to work with the fs that is the default. I have no scientific proof just 6th sense and experience with older distro versions.
Lately, I am thinking about having some kind of de-fragmentation, preferably with a GUI, just like in Windows NT defrag. I’ll have to google it sooner or later.
Kernel 2.6 better FS algorit … looks like the Autumn will be a good season for Linux enthusiast.
Lets not forget to expect some of the FreeBSD updates too.
You shouldn’t have to degrag any ext2,3/Reiser.
it doesnt fragment unless you’re running super low (<400MB) on space.
I use Reiser and it’s preformance is great for my needs alone. If my system goes down for whatever reason ( power-outages, etc..) Reiser does it’s checks in a hearbeat and you move on. Reiser works great for large directories filled with small ( I’d say 5-10mb and less in size ) files. Of course my needs are not much compared to a big time company. Anyways I am looking fowards towards the release of Reiser 4.
You don’t need to defrag ! It’s not neccessary at all in Linux and especially with the files systems like Reiser, XFS, JFS, ext2, ext3, etc…
// You don’t need to defrag ! It’s not neccessary at all in Linux and especially with the files systems like Reiser, XFS, JFS, ext2, ext3, etc… //
Hans Reiser himself said — at his LinuxTag speech about ReiserFs v4 being an atomic file system — that with ReiserFs v4 he will also release some sort of a defragment tool and cleared out why ReiserFs v4 will benefit from defragmentation or why it has to be defragmented (it’s an issue with the “dancing trees” being reordered for performace issues IIRC).
Damn, I thought I got away from that. I don’t need ANOTHER maitenance task.
Oh well, I guess the improved speed will make it worth it.
Do you know if you will be able to defrag mounted partition w/o problem? Because I don’t want to have to restart just to degrag.
// Do you know if you will be able to defrag mounted partition w/o problem? Because I don’t want to have to restart just to degrag. //
quote from Hans Reisers speech at LinuxTag:
The ReiserFs v4 Defragmentation Tool (along with other ReiserFs v4 tools that will be available e.g. a tool for online filesystem resize) will only work on _mounted_ partitions. Thats because the tools are actually only wrappers for accessing the functions implemented in the kernel module. The userspace tools do not reimplement any features for ReiserFs.
The defragmentation process for example will not actually move any bit of data instead the internal data tree will be newly calculate and/or balanced. As far as i understood Reisers speech only the metadata will be reorganized.
I attended Hans’ presentation at Linuxtag.
Basically, reiser4 is optimized for the case where you unpack a large tarball, say the Linux kernel, and have enough memory to hold it all in cache, which is true for most of us these days. reiser4 will then choose the optimal disk layout for these files and flush them to disk.
Hans also has aspects of a log structured file system in reiser4, which means you don’t write to the file, you write to a log file which basically encompasses the whole disk. The up side is that you mostly write linearly, the down side is that the files get badly fragmented if they are updated at all. Most files are not updated, just written once at installation of the package. The files that are updated frequently tend to be source code from CVS, which are small enough to fit in memory completely and have reiser4 choose an optimal disk layout again.
The case where this model completely sucks is the case where you update many portions of a large file. For example, running an SQL database with files on a reiser4 file system as backend, or maybe a DNS server with DDNS, or a berkeley db backend for Postfix or qmail to keep the SMTP AUTH users or something. Also, log files will probably be badly fragmented.
Hans proposes to have something like a transparent defragmenter running in the background, which he calls “repacker”. This would run in the kernel space, as part of the file system, and defragment badly fragmented files that are accessed frequently. This would solve most of the down sides of his approach, but this repacker is not finished yet.
My personal view of reiser4 is: it looks like it is optimized to perform well in benchmarks. It tries to be fastest for updating databases, but buys the performance by being slower when reading the data afterwards. The critical question is whether the repacker can alleviate these concerns, and as long as it is not finished, reiser4 is basically out of the question except for a little testing here and there. I reckon reiser4 would be a great filesystem for keeping your mozilla and gcc CVS checkout handy. But until the repacker is done, I will not even use it for testing, because the repacker really is the crucial component that makes or breaks this.
By the way: my previous experiences with reiserfs were less than stellar. Some people call it shredderfs instead. The main complaint with reiserfs is and always was that the fsck is not nearly as trustworthy or stable as the one from ext2/ext3. So even if I use reiserfs at all, it’s only for data I can afford to lose completely, like my CVS checkouts or the squid cache directories or something like that.
The benchmarks do look good though, and I am glad that at least someone is still trying major innovations in this area. Since most Unix vendors or divisions are no longer profit centers, file system innovations have largely stalled or moved to specialized companies who regard them as proprietary (Veritas) instead of releasing them as free software like IBM and SGI did.
for more about the defragmenter (Hans call it “repacker”) look at the documentation:
What is the license for ReiserFS? LGPL or GPL? I hope it is LGPL so that it can be added to operating systems with different licenses without infringing on the ReiserFS license.
“ReiserFS is hereby licensed under the GNU General
Public License version 2.”
“Further licensing options are available for commercial and/or other interests directly from Hans Reiser: email@example.com. If you interpret the GPL as not allowing those additional licensing options, you read it wrongly, and Richard Stallman agrees with me, when carefully read you can see that those restrictions on additional terms do not apply to the owner of the copyright, and my interpretation of this shall govern for this license.”
Soo, let me get this straight… not only do the fsck tools for reiserfs suck, not only is it slower then xfs, but he wants me to defrag the thing too?
I don’t get it, why would anyone use Reiser4 over XFS? someone please enlighten me…
is there possible to use it under rh 9?
any tools to convert ext3 to Reiser4?
Okay… to clarify.
The fsck tools are fine. I and many, many other people have never had any problems with a ReiserFS machine. In fact, ext2 has caused me to lose more data (welcome to non-journalling FSs).
It’s not slower than XFS in my experience although there are some areas for tweakage (which is why we would like Reiser 4, more than anything else).
You don’t have to Defrag the thing, it’s done transparently when the trees are rebalanced and in such a way as to optimise the locations of the data on the disk for maximum performance in future reads. It’s one of the performance-enhancing features.
> Linux doen’t run well on FAT/NTFS partitions.
If micro$ didnt sue the people who were making NTFS RW drivers then perhaps …
Erm is there a good easy disk formatting utility I can download and give it a try? Can I still install my favorite Linux distro on it, or do I need a particular distribution to run it?
What are the uses of or comparisons between IBM’s, SGI, Redhat’s and Reiser filesystem