Slashdot Mirror
Windows Alternatives to NTFS?
Maidjeurtam asks: "I'm a multi-OS user. Although Linux is what I use the most these days (I run it on my primary P4 box and on my iBook), I also run Mac OS X and a Windows XP on other machines. Of course, those boxes are networked, but sometimes, I just prefer to plug one machine's hard disk into another. I often work with big DV files (> 4GiB) and it looks like I have no other choice than having a different filesystem on each of my boxes. Granted, Linux can read NTFS (Macs can too) and even write to NTFS partitions thanks to tools like Captive, but I don't like the idea of running Windows code on my Linux box. In fact, I don't want my data stored on a proprietary, closed filesystem. I've googled a bit and it seems there's no modern (free-as-in-speech) filesystem I can install on Windows. I'd love to have ReiserFS running on my XP box, for example. Am I condemned to stay with NTFS, or do you guys know of a Windows-compatible, open filesystem that I can use?"
Last I checked, you couldn't have files over 4 gb in size on a FAT partition.
Maybe not, but most modern DV tools have options to break files into manageable-sized chunks, usually 1 or 2 Gb.
What we did for file systems was standardize on ISO9660 throughout, though we are considering a move to UFS so we have support for larger files. Windows supports it, though you have to hack the registry to get write access enabled, and we ended up writing a custom "disk format" tool to actually get disks initialised because, needless to say, W2K doesn't actually allow you to format disks as ISO9660 by default.
Well recommended. There are some neat features of ISO9660, like the VAX/VMS style "versions" for instance. Unfortunately, bog standard ISO9660 has crappy 32.3 style filenames (and for maximum compatability we're encouraged to just do 8.3), so it's not a perfect solution (another reason we want to switch to UFS.)
Definitely recommended though. It's a little slow, but everything will read it.
http://www.daniweb.com/techtalkforums/showthread.p hp?t=693
/* oops I accidentally made a comment, sorry */
And on top of that there is a ~32GB partition size limit.
That's an artificial limit imposed by Windows, to get people to switch to NTFS. Pick up a copy of PartitionMagic, it'll create those large FAT32 partitions just fine.
- Am I condemned to stay with NTFS?
Yes, as long as you want stability and consistenty (as in error recovery, such as provided by metadata journaling), you are.Windows supports FAT32 and ISO9660 out of the box. FAT32 does not provide enough error recovery to be recommendable. People using ISO9660 as a hard-drive file system are crazy masochists -- enough said. There are seemingly abandoned ports of Ext2 and ReiserFS out there. None of them are in any sense stable for production use.
Why aren't there more file systems available on Windows? The first clue is that Windows is not an open-source platform; open-source hackers tend to live on open-source platforms. The people who work on kernel-level development under Windows are likely to be pursuing commercial software from the outset.
Furthermore, Windows kernel development is something of a black art; it is hard enough that you need to have some vested interest in the platform in order to stay; you would want to live and breathe Windows kernel APIs. (APIs, incidentally, that don't seem constructed for use by humans; for example, due to the limited size of the kernel addressing space, there are several different "kinds" of memory you must carefully allocate and manage yourself. Add to this the awkwardness involved in debugging this stuff, the poor kernel-level development tools offered by Microsoft, the limited documentation, the fact that much third-party information is non-gratis, and of course that the kernel sources themselves are closed, and you have one painful hobby.) In short, you would want to become a kernel specialist.
These painfully-accrued skills are worth their weight in gold, and used to leverage careers as highly-paid consultants or highly-paid trainers, or both. And some, of course, are driver writers for hardware companies.
There's a further reason: Linux file system drivers, in my experience, are designed to be, well, Linux file system drivers. Witness the amount of effort taken by IBM and SGI to port their proprietary journaling file systems to Linux -- and this was from one Unix-like kernel to another. Windows' internal file-system driver API is completely different from Linux'. Porting one file system not only requires a lot of knowledge about the different kernel APIs, but also about the file system itself, because most likely the file-system code is not cleanly separated from the kernel-specific code; you can't just sit down and write an adapter layer. (This is actually mostly speculation, but based on casual perusal of some existing driver code.)
There will be viable, open-source file systems on Windows the day somebody takes the time and effort do implement (and maintain) one. As for myself, I bought the book and started; I gave up not because it was technically challenging, but because it was no fun, and there were more interesting knowledge out there that I wanted to store in my brain.
May work!!
/etc/nsswitch.conf
Its read-only, command-line access!
You type commands like this:
extdump \\.\PHYSICALDRIVE1 -o 68b3cb000
If you want to read a file.
Sam
blog.sam.liddicott.com
Many reasons:
1. The officially-sanctioned IFS developer kit is separate from the DDK and costs ~$1000 last time I checked. That's steep for someone who's hacking on open source in their spare time. Even if you get it, AIUI, it's very underdocumented and you'll take quite a few lumps before producing anything useful. And the license looks to my (untrained) eye as if it specifically prohibits distributing source code for the file system drivers you develop.
2. Windows filesystem drivers are hard. Developing and debugging is hairy and time consuming, and it's a good bit more work than writing an equivalent UNIX-like filesystem driver. For example, Windows expects the filesystem to handle globbing (wildcards), which is normally handled by the shell on UNIX.
3. Kernel drivers require specialized knowledge to develop and maintain. The people who have acquired this knowledge about the popular Free/Open Source drivers are, naturally, UNIX experts. They are unlikely to have the same knowledge about developing Windows kernel drivers and very unlikely to enjoy working with Windows enough to gain this knowledge in their spare time, at their own expense.
4. People who understand windows kernel driver (especially IFS) development don't, as a group, do open source. I have a few friends who do this, and they have actually mentioned that they view open source as a threat to their livelihoods and hope it goes away. One of them used the phrase "commie bullshit". I'm not joking.
5. The free kits for Windows filesystem development appear primarily targeted at academic use, and are not robust enough (or maybe simply not well-enough understood) to produce a production-quality filesystem. Some development would most likely need to go into one of the kits before it could be used to port a non-trivial filesystem with all of the expected features.
This all adds up to a serious lack of any "somebody" who can and is willing to write one, especially for free.
That said, there are a couple of free ext2/3 implementations available for various versions of windows. I've tested them, and the one that was read/write didn't seem good enough for use with any important data.
One company that I know of has developed a commercial implementation of ext2/3fs for Windows. It's not free, but for <$30 it may be inexpensive enough to be interesting.
.sig: file not found
Couple of things to check.
First, it sounds like you have two different partitions on the same hard drive. That's a no-no for benchmarks. The first partition (the one at the outer edge of the disk) will always have much better performance than the second partition (the one closer to the middle). The disk spins at a constant RPM, but the outer cylinders have more tracks on them, so you get more data per revolution.
Second, the 1k default "segment" size for NTFS (cluster, methinks) only kicks in for fairly small disks. It is an explicit tradeoff between throughput and space efficiency. By using 1k clusters, you get a lower proportion of wasted space, but you have to spend more effort tracking down all of the additional clusters. With 16k clusters, you get a higher transfer rate at the cost of more space wasted per file. With a sufficiently larget disk, NTFS defaults to 4k clusters, which is a good default for most people. NTFS performance doesn't increase much after 4k for the average workload, and due to "resident" data streams, this doesn't waste much disk space either.
In any case, my experience has been that the performance difference between NTFS and FAT (which has never seemed to be very much) is way less significant than the reliability and extra features offered.
Time flies like an arrow. Fruit flies like a banana.
Windows already has a local filesystem abstraction layer. File systems have drivers, like any device. NTFS (or any other fs) is not integrated in the kernel.
cdfs.sys for CDFS
fastfat.sys for FAT12/16/32
mrxdav.sys for WebDav (access files over http using normal file ops)
mrxsmb.sys for SMB
msfs.sys for mailslots (the mailslot filesystem)
mup.sys- the multiple UNC provider driver; to handle \\computername\share\path
npfs.sys for named pipes (the named pipe filesystem)
ntfs.sys for NTFS
udfs.sys for UDFS
Windows certainly supports third-party file system drivers. The problem is that the Microsoft IFS (installable filesystem) kit costs $899. There is a free/open alternative here.
However, in practice, there are no alternatives to MS filesystems. Your suggestion of a SMB server over a fast connection is a good one.
Quote (from Linux Kernel v.2.6.5 Configuration):
"CONFIG_NTFS_RW:
This enables the partial, but safe, write support in the NTFS driver.
The only supported operation is overwriting existing files, without
changing the file length. No file or directory creation, deletion or
renaming is possible. Note only non-resident files can be written to
so you may find that some very small files (
It is perfectly safe to say N here."