you want a lightweight interface? Try Inferno (see http://vitanuova.com./ Andrey Mirtchovski got the non-X11 (uses framebuffer) version working. Where a 128MB OLPC seems tight for memory, inferno happily runs its 120KB web browser, and it feels lost in 32 MB -- yep, we don't even have to bother giving it more than that.
You get a reasonable interface, you get a good language (Limbo -- far better than Python!), and you get a more modern OS universe than Linux.
Well worth a look -- now that I'm not working on the LinuxBIOS side any more (a full-time guy at OLPC now handles the LinuxBIOS for OLPC) I intend to do a native port of Inferno to OLPC.
BTW, if you want a command prompt, you can load Linux into the BIOS FLASH, and get an ash prompt on bootup -- or, in the system-as-shipped you can get a Forth prompt. Your choice. But getting to a command prompt on boot is easy, and you can then boot from a USB stick, and run whatever you want. You are NOT restricted to the shipped UI in any way, or even to Linux for that matter. Freedom to hack is a very important part of what OLPC is providing.
No more development eh? You don't know what you are talking about. DOE is funding more Plan 9 development. As of this year, we have a new Plan 9 C compiler for the 64-bit Opteron, we have the first-ever 64-bit Plan 9 kernel, for Opteron, and we got all the kernel cleanup that it required.
Lucent, I am told, uses Plan 9 quite a bit in embedded work. It makes money. It's not just a research OS. I expect development will continue.
Why not open boot? Because the "open" means only "open spec".
Open Boot is not Open Source Have you ever wondered why nobody ports it to lots of things? Or why http://www.openbios.org exists? Simple. Open Boot is a marketing name.
Again, Open Boot is NOT Open Source. It's just a cute name that seems to fool lots of people.
But go ahead, prove me wrong: point to the Open Source site for Open Boot.
it ought to be good enough for NASA./* pcnet32.c: An AMD PCnet32 ethernet driver for linux. *//*
* Copyright 1996-1999 Thomas Bogendoerfer
*
* Derived from the lance driver written 1993,1994,1995 by Donald Becker.
*
* Copyright 1993 United States Government as represented by the
* Director, National Security Agency.
*
* This software may be used and distributed according to the terms
* of the GNU General Public License, incorporated herein by reference.
*
* This driver is for PCnet32 and PCnetPCI based ethercards
*/
There are ca 24 files or so with this notice in drivers/net.
So the NSA seems to be ok with sending out stuff without requiring users to register at a web site. It seems possible to me that NASA could figure this one out.
Many US Gov't agencies now support creation of GPL'ed code, including DOE and DOD (via DARPA at least).
This license looks like another case of NASA Not Invented Here at work.
we're doing exactly this now on 1408 K8s, 256 K8s, 1100+ P4s, and several dozen Alphas here at LANL, and have been for years. Works fine. See http://www.linuxbios.org
And there is a GPL open boot firmware project at http://www.openbios.org.
This all works and is easy; it just needs people to help out.
Of course there are open blade servers. Probably the most interesting is http://www.cluster-labs.com/. 1 Ghz. PIII, fits into CompactPCI, runs LinuxBIOS (so my bias is obvious, right?).These blades come with all the source you need to change whatever you want. They're tiny, hot-plug, and really reliable.
A slightly less open blade, but still not bad is the RLX 800i, which also runs LinuxBIOS, although RLX is not officially supporting it.
Those "Open Blades" are out there, but journalists usually don't get much beyond Dell, Compaq, etc. so they always miss the interesting stuff. I guess it is too much work or the deadlines are too short to get a lot of information into these articles.
The LNXI modules in Pink are actually.8U effective space. I have a 1U RLX rack here with 800is. As much as I like the RLX, fact is that in terms of computing per cubic inch, the LNXI node is about 5 times the performance. Yup, 5 times:
The LNXI module takes up about the space of 4 of the RLX blades. In the space of four RLX PIII/800s, LNXI puts 2 2.4 Ghz. Xeons. On some of our apps, measured performance of 1 2.4 Ghz. Xeon is 2.8x the PIII/800. Do the math: same cubic inches, ca. 5x the throughput, 1/2 the number of CPUs.
Not being a blade or embedded node doesn't automatically mean you are less dense in computing per cubic inch!
I realize this is Slashdot:-), but I think at this point you need to backup those statements with one or two facts or citations. I can give you one, however: at several conferences in 2000, e.g. ALS 2000, Compaq demonstrated 16 and 32-node Alpha SMP systems with Linux. Scaling did stop at 16 (for kernel builds) with the version of Linux they ran back then. I had a 16-node GS160 and it did scale just fine to 16 nodes.
Can you actually provide a reference for that 32-node Windows box? Most of the "32 CPU Windows" boxes I have seen run Windows in cells of 4 CPUs, with 8 copies of the OS (e.g. Unisys). Do you really call this scaling? I don't.
Why not cPCI? In a word, performance/price on our apps. We looked at all sorts of cPCI blades (e.g. http://www.cluster-labs.de) but the peformance just is not there. Also, no existing ethernet will do the job for our apps, so we have to use Myrinet, and again, the fastest Myrinet is going to be in the PCI 64/66 slots on plain old motherboards.
Other folks have asked about 1.0 Ghz G4. I like PPC. But on our applications the PPC, with the best compilers we can find, is actually slower in absolute terms than a PIII/800. So as much as I would have wished to use a PPC, it's not cost-effective.
Note that our software runs fine on G3 and G4 macs however -- our standard CD distribution from http://www.clustermatic.org will boot either PC, PPC, or Alpha just fine. In fact, the standard Linux distribution from
http://www.terrasoftsolutions.com/products/blackla b/components.shtml
features some of our software, including bproc.
Also, if you look at the PPC offerings from synergy and CSPI you'll find they run their own kind of "Linux in flash" -- not LinuxBIOS, but pretty much the same function. They've been doing this for years.
HMMM, I'm sorry that you have failed to see how this is unique:-) You probably should visit http://www.clustermatic.org and read what's there.
Of course it has been going on a long time, I first did it 12 years ago with Suns. But your little 116-node cluster probably did not run into the problems you hit at a larger scale.
Anyway, what linuxbios gets us:
- more sane platform configuration
- we load linux from flash so can use all the
capabilities of linux as our bootstrap
- we boot over myrinet
- we're not even cabling the ethernet up
- We don't need to set up the serial network which
you HAVE to set up with kludges like SRM
You're just not going to get that with PXE or SRM.
I realize this detail was not available on the short article.
Why hasn't it taken off? because it doesn't boot windows, and I don't really care about that.
But if you are counting linuxbios is a ca. $16M business this year, so that's good enough for me. Plus, it runs in places you might be surprised to see...
see www.cluster-labs.de or linuxbios.com or lnxi.com or cwlinux.com for motherboards you can buy that run linuxbios.
In industry you try to get nodes for ca. $1200. That's how I used to build them anyway. You can't run NT Client on a cluster node -- you have the "10 TCP connections" limit for one thing, so MPI dies hard.
What's an NT server license cost with all the trimmings? Ca. $1200.
So you can build a cluster with Linux, or build a cluster with NT and watch half your money go to software cost.
Which is why NT will always be a loser for clusters.
Power Cockpit code is not GPL. A lot of other Turbo code is not GPL either. If Turbo goes, don't expect to ever see the source for these tools.
One thing Turbo got right -- it's hard to make a living selling free stuff. They wanted to make a living selling proprietary software for Linux -- the Linux part came free, but apps like Power Cockpit were licensed.
I use past tense, but given the lack of information, who knows what's going on right now.
Slashdot Mirror
you want a lightweight interface? Try Inferno (see http://vitanuova.com./ Andrey Mirtchovski got the non-X11 (uses framebuffer) version working. Where a 128MB OLPC seems tight for memory, inferno happily runs its 120KB web browser, and it feels lost in 32 MB -- yep, we don't even have to bother giving it more than that.
You get a reasonable interface, you get a good language (Limbo -- far better than Python!), and you get a more modern OS universe than Linux.
Well worth a look -- now that I'm not working on the LinuxBIOS side any more (a full-time guy at OLPC now handles the LinuxBIOS for OLPC) I intend to do a native port of Inferno to OLPC.
BTW, if you want a command prompt, you can load Linux into the BIOS FLASH, and get an ash prompt on bootup -- or, in the system-as-shipped you can get a Forth prompt. Your choice. But getting to a command prompt on boot is easy, and you can then boot from a USB stick, and run whatever you want. You are NOT restricted to the shipped UI in any way, or even to Linux for that matter. Freedom to hack is a very important part of what OLPC is providing.
thanks
ron
No more development eh? You don't know what you are talking about. DOE is funding more Plan 9 development. As of this year, we have a new Plan 9 C compiler for the 64-bit Opteron, we have the first-ever 64-bit Plan 9 kernel, for Opteron, and we got all the kernel cleanup that it required.
Lucent, I am told, uses Plan 9 quite a bit in embedded work. It makes money. It's not just a research OS. I expect development will continue.
thanks
ron
Words matter. When you call this stuff anti-piracy you're already surrendering the high ground to MPAA.
Please, if you're going to talk about this stuff, why not 'so-called anti-piracy', which is true, or, better yet, 'anti-fair-use'
thanks
ron
Why not open boot? Because the "open" means only "open spec".
Open Boot is not Open Source Have you ever wondered why nobody ports it to lots of things? Or why http://www.openbios.org exists? Simple. Open Boot is a marketing name.
Again, Open Boot is NOT Open Source. It's just a cute name that seems to fool lots of people.
But go ahead, prove me wrong: point to the Open Source site for Open Boot.
it ought to be good enough for NASA. /* pcnet32.c: An AMD PCnet32 ethernet driver for linux. */ /*
* Copyright 1996-1999 Thomas Bogendoerfer
*
* Derived from the lance driver written 1993,1994,1995 by Donald Becker.
*
* Copyright 1993 United States Government as represented by the
* Director, National Security Agency.
*
* This software may be used and distributed according to the terms
* of the GNU General Public License, incorporated herein by reference.
*
* This driver is for PCnet32 and PCnetPCI based ethercards
*/
There are ca 24 files or so with this notice in drivers/net.
So the NSA seems to be ok with sending out stuff without requiring users to register at a web site. It seems possible to me that NASA could figure this one out.
Many US Gov't agencies now support creation of GPL'ed code, including DOE and DOD (via DARPA at least).
This license looks like another case of NASA Not Invented Here at work.
ron
And there is a GPL open boot firmware project at http://www.openbios.org.
This all works and is easy; it just needs people to help out.
ron
A slightly less open blade, but still not bad is the RLX 800i, which also runs LinuxBIOS, although RLX is not officially supporting it.
Those "Open Blades" are out there, but journalists usually don't get much beyond Dell, Compaq, etc. so they always miss the interesting stuff. I guess it is too much work or the deadlines are too short to get a lot of information into these articles.
ron
The LNXI modules in Pink are actually .8U effective space. I have a 1U RLX rack here with 800is. As much as I like the RLX, fact is that in terms of computing per cubic inch, the LNXI node is about 5 times the performance. Yup, 5 times:
The LNXI module takes up about the space of 4 of the RLX blades. In the space of four RLX PIII/800s, LNXI puts 2 2.4 Ghz. Xeons. On some of our apps, measured performance of 1 2.4 Ghz. Xeon is 2.8x the PIII/800. Do the math: same cubic inches, ca. 5x the throughput, 1/2 the number of CPUs.
Not being a blade or embedded node doesn't automatically mean you are less dense in computing per cubic inch!
ron, one of the Pink guys
I realize this is Slashdot
Can you actually provide a reference for that 32-node Windows box? Most of the "32 CPU Windows" boxes I have seen run Windows in cells of 4 CPUs, with 8 copies of the OS (e.g. Unisys). Do you really call this scaling? I don't.
ron
yep, you can do it with floppies. But do you really want to do it with 1024 floppies given a 10% average failure rate? Think about it.
we run out of 256KB if needed. The 8 MB is for platforms (http://www.cwlinux.com) where people want a local file system in flash.
We had 104 DS10s booting out of 2 MB flash no problem. So there are variations.
Too bad about DBLX (your company's product) but that also speaks volumes about that product, I guess.
Why not cPCI? In a word, performance/price on our apps. We looked at all sorts of cPCI blades (e.g. http://www.cluster-labs.de) but the peformance just is not there. Also, no existing ethernet will do the job for our apps, so we have to use Myrinet, and again, the fastest Myrinet is going to be in the PCI 64/66 slots on plain old motherboards.
Other folks have asked about 1.0 Ghz G4. I like PPC. But on our applications the PPC, with the best compilers we can find, is actually slower in absolute terms than a PIII/800. So as much as I would have wished to use a PPC, it's not cost-effective.
Note that our software runs fine on G3 and G4 macs however -- our standard CD distribution from http://www.clustermatic.org will boot either PC, PPC, or Alpha just fine. In fact, the standard Linux distribution from http://www.terrasoftsolutions.com/products/blackla b/components.shtml
features some of our software, including bproc.
Also, if you look at the PPC offerings from synergy and CSPI you'll find they run their own kind of "Linux in flash" -- not LinuxBIOS, but pretty much the same function. They've been doing this for years.
HMMM, I'm sorry that you have failed to see how this is unique :-) You probably should visit http://www.clustermatic.org and read what's there.
Of course it has been going on a long time, I first did it 12 years ago with Suns. But your little 116-node cluster probably did not run into the problems you hit at a larger scale.
Anyway, what linuxbios gets us:
- more sane platform configuration
- we load linux from flash so can use all the
capabilities of linux as our bootstrap
- we boot over myrinet
- we're not even cabling the ethernet up
- We don't need to set up the serial network which
you HAVE to set up with kludges like SRM
You're just not going to get that with PXE or SRM.
I realize this detail was not available on the short article.
Why hasn't it taken off? because it doesn't boot windows, and I don't really care about that.
...
But if you are counting linuxbios is a ca. $16M business this year, so that's good enough for me. Plus, it runs in places you might be surprised to see
see www.cluster-labs.de or linuxbios.com or lnxi.com or cwlinux.com for motherboards you can buy that run linuxbios.
Ron, the linuxbios guy
What's an NT server license cost with all the trimmings? Ca. $1200.
So you can build a cluster with Linux, or build a cluster with NT and watch half your money go to software cost.
Which is why NT will always be a loser for clusters.
Power Cockpit code is not GPL. A lot of other Turbo code is not GPL either. If Turbo goes, don't expect to ever see the source for these tools. One thing Turbo got right -- it's hard to make a living selling free stuff. They wanted to make a living selling proprietary software for Linux -- the Linux part came free, but apps like Power Cockpit were licensed. I use past tense, but given the lack of information, who knows what's going on right now.