posted by aldie_lab

posted by Charles

posted by felix9

This means that we might finally get native code on the web(again)?

posted by philjay

posted by Simon

If app consists of multiple sequential or parallel executables, so that eg. excel.exe starts excel2.exe and then excel.exe terminates and excel2 starts multiple different exes with their own windows and excel2.exe terminates... will this kind of thing work with this model? What if there's also some LPC or shared memory IPC between these before the termination?

If app uses CreateFile to open \\.\C: (hope i got that right) or a PhysicalDisk and in order to run needs to be able to write and read somewhere on the disk without going through the filesystem apis, will your security layer virtualize this or will the app fail to run?

How do you "install" app onto this sandbox? Lot of talk about lack of 3D/HW support but would have been many more interesting questions about how to handle things related to what eg. game installers do, such as "sony rootkit drm", would that rootkit drm game install fine even if it was just 2D non-accelerated game. Also, would this approach work to enable better compatibility with Windows 3 & 95/98 apps/games using old DX apis?

Getting old windows games and apps to run is oft more pain than dos games in dosbox. If MS were to productize this research, it could end up like the current app compat layer, which can require a bunch (too much) of fiddling just to find the app you want to run is not going to run since even if you put compat mode "XP", the broken stuff tends to stay broken unless it was specifically tested by people in MS.

I think this type of legacy compatibility thing may be better using a hybrid development model: paid core team developing the long term goal deliveries and then allow the community using the product develop their own minor fixes and improvements that could be easily patched (by users, so simply that no instructions are needed) into the product on need basis. eg. if I as user run appX, it will check for community made fixes for appX and allow me to install those in the sandboxing layer or something, ensuring longevity and broadening compatibility as time goes on even if MS stops active development on the sandbox. Just a thought...

posted by androidi

C

posted by Charles

posted by Minh

posted by Charles

Not to mention the possibilities as far as backward compatibility is concerned.

posted by rstat1

posted by felix9

16 hours ago, Charles wrote

@Minh:Weird. Republishing.
C

Still not working here either. other videos work fine...

posted by giovanni

posted by Charles

posted by staceyw

11 minutes ago, staceyw wrote

Nice. So in the future, a user could hit a exception, then "click-dump" the process (as a button in the exception window) and email to me. I could open that in VS debuging and be right in the context of the issue and even see what happened before the exception.

If you want to resume hibernation, you need hiberfil.sys as well as your intact filesystem. You can't just send hiberfil.sys to another machine and resume your OS there.

posted by JohnSawyer

C

posted by Charles

http://technet.microsoft.com/en-us/appvirtualization/dd146065

The move from kernel mode to user mode comments stood out.

Very cool stuff!

I'm curious how (in conceptual terms) the Drawbridge compares to a technology like Thinapp (previously Thinstall), I suppose other than the obvious ability to rearrange the OS..

posted by Scottee

This keeps memory usage low. How many instances of IIS would you be able to run in VMs using memory deduplication, as opposed to the number in Drawbridge? How well does Drawbridge perform CPU-wise, as opposed to running on bare metal hypervisor?

There were also some scenarios mentioned, such as:
- using it to keep compatibility with XP
- sandboxing

Well, I don't think it is easy to refactor an outdated OS and to keep compatibility for every single system call. Would you use XP RTM, XP SP1, XP SP2 or XP SP3 as the baseline?

With an upgrade to a new OS version, your existing applications get a new look, since they blend with the OS's redesigned UI elements. This is very much desirable, as opposed to keeping it at the version "they were designed for". Then there's WinRT. I guess "desktop mode" APIs will stay Win7 compatible for a very long time, since most innovation will be in the WinRT world.

As to sandboxing, using a processor security feature (ie VM mode) is much more secure than it is to use existing ring protection. Unless you decide to use PL1 and PL2 (how's with ARM compatibility then)?

posted by JohnSawyer

posted by martinmine

20 hours ago

Linux Containers (http://lxc.sourceforge.net/) anyone?

@James: As you point out, there is a rich history of sandboxing technologies that operate at the scale of an application (chroot, zones, jails, containers, etc.). These were all important advances. Our contribution is to marry application sandboxing with the library OS concept. If you want to read more detail, our ASPLOS 2011 paper provides some comparison with existing technologies.

As far as we know, Drawbridge is the first in this class to provide not just isolation, but also persistent compatibility and execution continuity. When packaged with its library OS, a Drawbridge application can run across many different host OS versions.  And, a running Drawbridge application can move from one host machine to another (without losing its state).

posted by galen

posted by giovanni

I'm jealous of their lifestyle.

posted by B3NT

Oct 19, 2011 at 12:47 AM, JohnSawyer wrote

I watched the video yesterday. But it got me thinking - how exactly is rearranging the OS better than using a Hyper-V VM with memory deduplication on EPT/Nested Pages?

This keeps memory usage low. How many instances of IIS would you be able to run in VMs using memory deduplication, as opposed to the number in Drawbridge? How well does Drawbridge perform CPU-wise, as opposed to running on bare metal hypervisor?

Any benchmarks yet?

posted by JohnSawyer

Oct 30, 2011 at 2:05 PM

In the paper "Exterminate All Operating System Abstractions" (www.stanford.edu/~engler/hotos-jeremiad.ps) they talk about an "application-level operating system"; would you say that that is, or can be seen as, related to DrawBridge?

Yes, Engler et. al invented the idea of a library OS (an "application-level operating system").  Our academic contribution was to show 1) how the interface between the library OS and the host OS can be modified to enable persistent compatibility, 2) how the it can enable migration, and 3) that Windows can be used to create a library OS.

By the way, our paper mentioned above discusses the related work in more detail.

posted by galen

Oct 17, 2011 at 12:27 PM

Unrelated: Would love to see Singularity's ideas go mainstream. Singularity's *the single most interesting* idea and implementation I've seen in a long time in OS research. While it might not have been the first approach it was very well executed (and documented).

Thanks!  We are very proud of our Singularity work as well.  Interesting, several of the great ideas from Singularity were reused in Drawbridge.  For example, the Drawbridge ABI (application binary interface) is very similar to the Singularity ABI.  Also, Drawbridge employs many of the program manifest and packaging ideas that we pioneered in Drawbridge.

posted by galen

posted by ajasmin