Isn't latency more important for common desktop tasks than throughput?
Like the whole Control Panel is all GUI based, is there some way to do that configuration from a CLI? It might be possible to do in the registry, but the whole registry is a screwed up mess. Is there even a CLI registry browser in Windows. Regedit.exe is a GUI program. :/
I do know that you can edit the registry through Batch, but you have to know ahead of time what keys you're editing. I'm not aware of a browser like Regedit where you can just peruse the registry.
Another problem with running Windows without a GUI is that it'd be a pain in the ass to install software, not just because most programs require a GUI, but because there isn't a package manager and there isn't a built-in wget/curl utility for downloading executables. So you would use another computer and a browser to download the file then transfer it to your server.
The standard CMD shell just seems like it's lacking too much to be the sole interface to a system.
It's easy to make a file system fast when you don't care about your user's data.
hmm first here it goes about a filesystem. Nearly all of you just belived that advertisement for a commercial product, before I see some independent tests I think its even a waste of time to talk about it.
The talk about windows vs linux is funny. I got critesised in this forum when I did say: "its useless because its proprietary closed source stuff", because then I would speak for all linux users.
But basicly the most import stuff, why linux and other free/open software/os is better, because its free ^^, thats the main reason. Windows for desktop is good enough for most tasks, you can argue with gnome-shell or something, or for development windows also sucks, but there is nothing some propriatary software can not even out at least to a near point, and some stuff is even better on windows. (opengl seems to be stuck and dx11 seems to be better technicaly) But thats the point, if you care about open source there is no reason to defend your os/software of your choice or attack a blob some other people like to use.
Its just if you want to have the 4 freedoms or you want to be a begging mightyless guy that have to bless for wonders of your masters (developers/ms...)
If you want to give away stuff or copy stuff without getting accused to be a criminial on a ship (pirate). And security because it got so much quoted before, In the windows world you dont know about it, because only one company knows, what they did write in the code. so bugs are in both worlds, but willful included anti-features are only possible in windows or other propriatary code blobs. and we know it did happen several times and sometimes we did not notice it and it is in there anyway because wihtout the code you cant surely find all this anti-features. As example the sony-rootkit.
So yes its also possible in opensource code, the change that nobody with good intentions did fly over the code is very low, or you can do it yourself on smaller stuff that not so many people use.
Last edited by blackiwid; 06-24-2011 at 05:41 PM.
Why didn't they choose Ext4? It doesn't fit, because people will notice where the catch is?A Phoronix reader has pointed out that a developer at Tuxera is claiming their proprietary NTFS Linux kernel driver makes the Microsoft file-system the fastest choice under Linux.
Benchmarks or it didn't happen.
Also, an embedded 800 MHz system? Seriously? You use that as a basis for showcasing this FS driver? It's probably ARM, too.
Useful for cellphones? Maybe, if Anton's initial embedded data are reliable. Useful for quad-core desktops with 6 GB of RAM? To be proven with benchmarks. Useful for high-end computing on quad-processor Xeon HexaCore servers with 64GB of RAM and 10kRPM SAS disks in RAID10? Extremely doubtful.
It's easy to write a driver that "outperforms" some other driver by either (a) sacrificing data integrity, or (b) optimizing for a particular workload (small systems, big systems, single-core, multi-core, etc). But what makes a truly great filesystem is one that performs reasonably well on all systems, and scales well from an embedded SoC all the way up to your 256-processor supercomputer. The focus with ext4 has always been on scalability, not cutting corners to be fast on an embedded system.
Also, the delayed metadata committing sounds dangerous. Every second you don't commit that data to disk is a second that the user's files are at risk. If you don't completely commit the journal, the metadata and the file data to the media when userspace calls fsync(), you aren't doing your job as a filesystem. I can't read the code, since it's proprietary, so maybe Tuxera NTFS does respect fsync() semantics, but my guess is that if they're going to cut corners somewhere to get higher performance, it would be there.
Exactly the same kind of thing that caused VirtualBox to have extremely high performance in some scenarios. Basically using a RAM-based filesystem in lieu of actually committing things to disk like the user asked.
Technical issues aside, the proprietary aspect of this would ensure that I am no more likely to try this product than KQ Infotech's ZFS driver (which I've never installed, even after it went open source).
Last edited by allquixotic; 06-24-2011 at 06:30 PM.
Since Microsoft Windows is still the dominating desktop OS and it supports a rather limited set of file systems of which FAT's 4GB file size limit can be problematic for content in HD, the choice to also add ntfs support is easy.