This doesn't make any sense. I'm running Solaris 10 on an Epia Mini-ITX system. You only get "below" that if you want to run something on the embedded CPU in your car. The exact same OS runs on a 72-way multi-threaded server with 1/2 TB of memory. There are machines larger than that, but it's not a "growth" part of the industry.
And in regard to Solaris on x86, it's completely senseless, for more than one reason.
Which makes it even more astonishing that it is so hard to get out of it. No Ctrl-Alt-Backspace to zap the X-server.
This is available if you use the Xorg server instead of Xsun. I thought Xorg was the default in s10? If ctrl-alt-backspace isn't working, try using the crtl and alt on the right side of the keyboard. I don't know why those are different than the equivalents on the left side, but they seem to be a bit more reliable in this situation.
You have to manually rifle through your 5 CD's, copy the package files to/var/spool/pkg, and run pkgadd manually
You don't have to copy them anywhere. Either "pkgadd -d/cdrom/..." and select your package from the list, or "pkgadd -d/cdrom/.../package_name".
How do you mount an USB keyfob, or similar device?
In theory, I don't think you should have to mount those at all. vold should do that automatically - just like it does for cds. In practice, getting S10 to recognize my iPod wasn't quite that easy. I haven't tried a USB device, so I can't say whether it will really work.
If vold doesn't automatically mount the keyfob, then try rebooting with the it inserted. Once it has been recognized once at boot, it should be recognized automatically in the future.
Oh, and telneting in from another machine is not an option, if your network card is one of the many that aren't supported out of the box.
On a real PC, you can often redirect the console to a serial line and use "tip" (or some Linux equivalent) to get to the machine's console. That also gives you a way to get a network driver onto the machine without burning it to a CD. uuencode it to ascii, and then use ~> to copy the file over. Since console redirection often isn't available on laptops, this may not work for you.
You can also try PXE booting your machine. Since the boot/install image is on a server, you can easily insert your driver into the image so it is available at install time.
As I understand it, they work almost identically to a SMP setup, meaning they don't offer much of any performance benefit in most apps (particularly games).
I suppose if your idea of "most apps" is games, then this probably isn't an area that would be of interest to you
If you have any multithreaded app that is even remotely competently written, then it will benefit from dual cores or (possibly) hyperthreading. If your multithreaded app is full of "big locks", then dual cores won't help, and the application designer is a failure.
If you have a workload that has multiple processes running simultaneously, then it is also likely to benefit from dual core. It gets more interesting with business/server workloads, but "home users" can benefit two. Even something as simple as running xmms and gcc at the same time should go faster. Or running two instances of lame.
The real win with dual core comes from increased throughput. A single job/application/process isn't likely to go any faster, but a full workload of multiple, reasonably parallelizeable, tasks will be faster.
Now, also, how many OSs (and applications) are prepared for dual-core support? Are there any available systems that are stable and do that?
Solaris supports dual cores on both SPARC and x86. The UltraSPARC IV processors are dual core.
Any application should be "prepared for dual-core support". If the application even has to be aware that it's running on a dual core or hyperthreaded CPU, then the OS is broken.
Jeebus, you're an idiot. Nobody said anything about stopping all construction or all use of wood.
There are some trees that probably shouldn't be cut down and some places where we shouldn't be building new houses. That doesn't mean we have to revert to the stone age.
But hell, I guess that's what passes for rational argument among right wingers these days. Bush has people lie about inconvenient facts. Since his mindless followers don't have that kind of power, they resort to building strawmen to tear down.
I was able to install a recent build of Solaris 10 on it without a hitch, so the functionality seems to be very solid. However, the installation took almost 6 hours, or about 10 times longer than a native installation. Since installation is all about I/O, this doesn't bode well for actually running the OS when the CPU performance will be much more important.
As for your suggestion that QEMU is similar to Xen: no, it's not. QEMU emulates the entire machine - including the CPU. This leads to the hideous performance I described. Xen doesn't emulate the CPU - the real CPU executes the instructions. I haven't used Xen yet, but this should allow it to run at near-native performance.
If one of these processors can compete with a 32 processor Sun Fire 12K in a functional sense, do you think Oracle is going to want less money?
It can't compete with the 12K in _any_ sense. It has a tiny fraction of the memory capacity, the memory bandwidth, the I/O bandwidth, the cache, the TLB, the floating point performance, etc.
This might be a great CPU for first-tier web serving, and maybe second-tier app-server kinds of processing, but it's not even a remotely viable candidate for the third-tier database work that Oracle cares about.
The IBM Open chip is not a chip without an OS. It runs linux...a commodity OS. That means two major things.
I read that to mean that it's a chip that doesn't have an OS 'based' on it. Linux runs on it, but Linux's base hardware is x86. Most of the Linux developers do their work, testing, and tuning on x86, so that's where it will run best. It will run on many other platforms, but they aren't going to get the same level of attention.
Similarly, Darwin is native to PPC. There is an x86 port, but the PPC implementation gets most of the work. Same with Solaris and SPARC (although, it seems like Sun is in the process of making both SPARC and x86 "native platforms" starting with Solaris 10).
IBM's new processor doesn't have an OS that is native to it. AIX doesn't run on it at all. Linux will run on it, but I can't imagine it will get the same attention as x86, amd64, PPC, or even MIPS.
What I don't get: if you want a low-end Linux machine, why wouldn't you buy an x86 or amd64 box? What is the incentive to buy a proprietary hardware platform just to run an open source operating system? It's bizarre. I'm with Jonathan on this: I just don't understand what IBM is thinking.
People not happy with big blue can migrate to another vendor without having to take an OS change into account. That means less lock in.
This doesn't follow at all. Nobody is going to buy these Power machines to run Open Source software. They'll run an Open Source OS (OS^2?), but they'll be running IBM's proprietary software. If you want to run Apache and/or MySQL, you'll go with cheaper, easier-to-find, and easier-to-maintain x86/amd64 hardware. If you want to run WebSphere, maybe you'll look at these low-end Power machines. Once you've done that, you are completely locked in: hardware, OS, and application.
Done (or at least getting there). Next time you're on a Solaris box, look in/usr/sfw/bin. Solaris now ships with bash (in/usr/bin) and GNU tar, grep, wget, texinfo, gs, ncftp (OK, not GNU but still usefull), and mozilla.
That's not insane, but it's probably not optimal either.:)
Hacking glibc to call the Solaris system calls (including all the necessary argument twiddling) would work in many cases, but it seems like a fragile solution:
Every time a glibc bug was found, Sun would have to reissue a patched glibc.
Statically compiled apps would never work.
System tools (like truss and dtrace) would report what glibc was doing instead of what the application was doing.
The only way I can see this working as a production-quality feature is if they actually implement the int80 mechanism and the Linux system calls in the Solaris kernel.
How do they handle the differences in threading models between Solaris ("real" threads) and Linux ("clone"ed processes that share address spaces)? Not a clue, but I'm sure it can't be done in a hacked up glibc.
Just as a simple, grade A, introductory issue: How does a Linux application issue a system call? Using int80. How does a Solaris application issue a system call? Using syscall, sysenter, or lcall depending on the application and the version of the OS.
The two OSes don't even agree on the basic mechanism by which applications can communicate with the kernel. And you think it's just a matter of putting glibc on the CD. Put down the keyboard and go back to CS101 until you learn something.
Sun and Fujitsu just announced a 20-year partnership to jointly develop SPARC based technology and systems.
They have had a partnership for 20 years - they aren't announcing a new one.
My guess: Sun drops Ultrasparc III
Sun is already shipping the Ultrasparc IV. Nice guess!
to provide the Throughput computing chips for the low end / web / network stuff,
They have already announced that this is exactly what they are going to do. Again: nice guess!
and takes up the Fujitsu provided SPARC64 chips for the high end and workstation market.
Yesterday's announcement was all about using SPARC64 on the high end. Usually the trick is reading between the lines - not reading the lines themselves.
Sun also announced that they will be using Opterons in their new workstation line - not SPARC64.
Will this spark a new RISC renaissance for Sun and Fujitsu? Or is it a last gasp before Opteron / PowerPC / Itanium crush them?
Itanium has gone white dwarf. The only thing it will be crushing is itself.
Opteron is not going to crush Sun. They have announced that they are shipping multiple Opteron boxes (1-8 way servers and 1-2 way workstations).
This could really revitalize the SPARC system market, especially if Sun's work on Throughput computing proves out.
This doesn't even make sense. The Fujitsu/Sun machines are not the Throughput Computing systems that Sun has been talking about for months. Throughput compututing is Niagara/Rock - the Sun-only CPUs.
You can keep saying that, but that won't make it true. They have the same name, the same functionality, and essentially the same interface. The mini IPod is a little bit smaller than a maxi IPod, but the difference isn't that significant. It's not like it's 1/2 the size.
I don't understand how they think the mini IPod will sell when a "real" Ipod is only $50 more. But, then, I'm just an engineer:). I'm sure any one of Apple's marketing people knows more about what people will spend than all of/. combined.
The AMD chip, on the other hand, can support x86 relatively easily by including a "morphing layer" (I think that's the name) which maps x86 instructions to the native instructions of the chip
This is only correct if you consider microcode to be the "native instructions" on AMD. Itanic introduced a whole new ISA, which I guess requires some kind of 'morphing to support x86. Opteron uses the existing x86 ISA with a small number of 64-bit extensions. So, x86 is the "native instruction" set for the AMD CPUs, which allows for much better performance of current 32-bit applications.
Is it just me, or does this thing sound exactly like the Cobalt Cube from a few years back
It _is_ the Cobalt Qube software. Jesus, I know nobody reads the articles, but is it really too much to ask that you read the title of the message you are responding to?
I certainly wouldn't pay $10/hour if I were planning to be online for many hours. However, if I have a one hour layover in Chicago on a BOS->SFO flight, I might pay $10 to get online for that one hour. If you charge $10 for the first hour and $1 for subsequent hours, then you start making more sense.
I would think that Sun is giving away the x86 Solaris for free because they just want to draw more users from the open source and free-as-in-speech community to look at what it has to offer...
They are re-releasing x86 support because they're customers were screaming for it. My guess is that they are giving it away to individual users because it's cheaper than trying to deal with $20 "micropayments.":) These guys are used to seeing money come in in 6-8 figure chunks. I doubt they are staffed to handle lots of tiny little transactions.
Much more hardware works than is listed on the HCL.
Some things aren't supported out of the box, like the ethernet controller (VIA Rhine) built into my motherboard, but there are a surprising number of third-party drivers available online. Wheel mice are also supported by non-Sun drivers.
One real shortcoming is in the area of video drivers. Fortunately, XFree86 works on Solaris x86, addressing that problem rather nicely.
The biggest problem I've run into thus far is a lack of support for Serial ATA. I expect that this is just a result of Solaris x86 being put on hold for a year or so. I can't imagine this wouldn't be available by Solaris 10 (or whatever they end up calling it).
There probably isn't a whole lot of reason to not choose Intel, other than the fact that Sun operates on religious principles rather than business ones.
Opteron is cheaper, faster, and requires less power than Itanium. AMD's 64-bit ISA is a hell of a lot cleaner than Itanic's. And AMD appears to be making the switch to 90nm technology faster and better than Intel, which will further help their power/cooling story.
Slashdot Mirror
all the growth is above them and below them
This doesn't make any sense. I'm running Solaris 10 on an Epia Mini-ITX system. You only get "below" that if you want to run something on the embedded CPU in your car. The exact same OS runs on a 72-way multi-threaded server with 1/2 TB of memory. There are machines larger than that, but it's not a "growth" part of the industry.
And in regard to Solaris on x86, it's completely senseless, for more than one reason.
OK, let's hear at least one then.
This is interesting. How exactly would you set it up?
8 g?a=view
This is one place you really do need to read the docs: http://docs.sun.com/app/docs/doc/817-5504/6mkv4nh
Which makes it even more astonishing that it is so hard to get out of it. No Ctrl-Alt-Backspace to zap the X-server.
This is available if you use the Xorg server instead of Xsun. I thought Xorg was the default in s10? If ctrl-alt-backspace isn't working, try using the crtl and alt on the right side of the keyboard. I don't know why those are different than the equivalents on the left side, but they seem to be a bit more reliable in this situation.
You have to manually rifle through your 5 CD's, copy the package files to /var/spool/pkg, and run pkgadd manually
/cdrom/..." and select your package from the list, or "pkgadd -d /cdrom/.../package_name".
You don't have to copy them anywhere. Either "pkgadd -d
How do you mount an USB keyfob, or similar device?
In theory, I don't think you should have to mount those at all. vold should do that automatically - just like it does for cds. In practice, getting S10 to recognize my iPod wasn't quite that easy. I haven't tried a USB device, so I can't say whether it will really work.
If vold doesn't automatically mount the keyfob, then try rebooting with the it inserted. Once it has been recognized once at boot, it should be recognized automatically in the future.
Oh, and telneting in from another machine is not an option, if your network card is one of the many that aren't supported out of the box.
On a real PC, you can often redirect the console to a serial line and use "tip" (or some Linux equivalent) to get to the machine's console. That also gives you a way to get a network driver onto the machine without burning it to a CD. uuencode it to ascii, and then use ~> to copy the file over. Since console redirection often isn't available on laptops, this may not work for you.
You can also try PXE booting your machine. Since the boot/install image is on a server, you can easily insert your driver into the image so it is available at install time.
As I understand it, they work almost identically to a SMP setup, meaning they don't offer much of any performance benefit in most apps (particularly games).
I suppose if your idea of "most apps" is games, then this probably isn't an area that would be of interest to you
If you have any multithreaded app that is even remotely competently written, then it will benefit from dual cores or (possibly) hyperthreading. If your multithreaded app is full of "big locks", then dual cores won't help, and the application designer is a failure.
If you have a workload that has multiple processes running simultaneously, then it is also likely to benefit from dual core. It gets more interesting with business/server workloads, but "home users" can benefit two. Even something as simple as running xmms and gcc at the same time should go faster. Or running two instances of lame.
The real win with dual core comes from increased throughput. A single job/application/process isn't likely to go any faster, but a full workload of multiple, reasonably parallelizeable, tasks will be faster.
Now, also, how many OSs (and applications) are prepared for dual-core support? Are there any available systems that are stable and do that?
Solaris supports dual cores on both SPARC and x86. The UltraSPARC IV processors are dual core.
Any application should be "prepared for dual-core support". If the application even has to be aware that it's running on a dual core or hyperthreaded CPU, then the OS is broken.
Jeebus, you're an idiot. Nobody said anything about stopping all construction or all use of wood.
There are some trees that probably shouldn't be cut down and some places where we shouldn't be building new houses. That doesn't mean we have to revert to the stone age.
But hell, I guess that's what passes for rational argument among right wingers these days. Bush has people lie about inconvenient facts. Since his mindless followers don't have that kind of power, they resort to building strawmen to tear down.
If you read the USENIX paper by the Solaris engineers who actually developed Containers, you will see that they talk about both Jails and VServer.
Performance numbers? No. Performance perception: unusably slow.
I was able to install a recent build of Solaris 10 on it without a hitch, so the functionality seems to be very solid. However, the installation took almost 6 hours, or about 10 times longer than a native installation. Since installation is all about I/O, this doesn't bode well for actually running the OS when the CPU performance will be much more important.
As for your suggestion that QEMU is similar to Xen: no, it's not. QEMU emulates the entire machine - including the CPU. This leads to the hideous performance I described. Xen doesn't emulate the CPU - the real CPU executes the instructions. I haven't used Xen yet, but this should allow it to run at near-native performance.
Have you priced Office 2003 lately? Absolutely REDICULOUS pricing model MS has.
Yes, but it comes with a spell checker.
If one of these processors can compete with a 32 processor Sun Fire 12K in a functional sense, do you think Oracle is going to want less money?
It can't compete with the 12K in _any_ sense. It has a tiny fraction of the memory capacity, the memory bandwidth, the I/O bandwidth, the cache, the TLB, the floating point performance, etc.
This might be a great CPU for first-tier web serving, and maybe second-tier app-server kinds of processing, but it's not even a remotely viable candidate for the third-tier database work that Oracle cares about.
The IBM Open chip is not a chip without an OS. It runs linux...a commodity OS. That means two major things.
I read that to mean that it's a chip that doesn't have an OS 'based' on it. Linux runs on it, but Linux's base hardware is x86. Most of the Linux developers do their work, testing, and tuning on x86, so that's where it will run best. It will run on many other platforms, but they aren't going to get the same level of attention.
Similarly, Darwin is native to PPC. There is an x86 port, but the PPC implementation gets most of the work. Same with Solaris and SPARC (although, it seems like Sun is in the process of making both SPARC and x86 "native platforms" starting with Solaris 10).
IBM's new processor doesn't have an OS that is native to it. AIX doesn't run on it at all. Linux will run on it, but I can't imagine it will get the same attention as x86, amd64, PPC, or even MIPS.
What I don't get: if you want a low-end Linux machine, why wouldn't you buy an x86 or amd64 box? What is the incentive to buy a proprietary hardware platform just to run an open source operating system? It's bizarre. I'm with Jonathan on this: I just don't understand what IBM is thinking.
People not happy with big blue can migrate to another vendor without having to take an OS change into account. That means less lock in.
This doesn't follow at all. Nobody is going to buy these Power machines to run Open Source software. They'll run an Open Source OS (OS^2?), but they'll be running IBM's proprietary software. If you want to run Apache and/or MySQL, you'll go with cheaper, easier-to-find, and easier-to-maintain x86/amd64 hardware. If you want to run WebSphere, maybe you'll look at these low-end Power machines. Once you've done that, you are completely locked in: hardware, OS, and application.
Done (or at least getting there). Next time you're on a Solaris box, look in /usr/sfw/bin. Solaris now ships with bash (in /usr/bin) and GNU tar, grep, wget, texinfo, gs, ncftp (OK, not GNU but still usefull), and mozilla.
Hacking glibc to call the Solaris system calls (including all the necessary argument twiddling) would work in many cases, but it seems like a fragile solution:
Every time a glibc bug was found, Sun would have to reissue a patched glibc.
Statically compiled apps would never work.
System tools (like truss and dtrace) would report what glibc was doing instead of what the application was doing.
The only way I can see this working as a production-quality feature is if they actually implement the int80 mechanism and the Linux system calls in the Solaris kernel.
How do they handle the differences in threading models between Solaris ("real" threads) and Linux ("clone"ed processes that share address spaces)? Not a clue, but I'm sure it can't be done in a hacked up glibc.
You're an idiot.
Just as a simple, grade A, introductory issue: How does a Linux application issue a system call? Using int80. How does a Solaris application issue a system call? Using syscall, sysenter, or lcall depending on the application and the version of the OS.
The two OSes don't even agree on the basic mechanism by which applications can communicate with the kernel. And you think it's just a matter of putting glibc on the CD. Put down the keyboard and go back to CS101 until you learn something.
Sun and Fujitsu just announced a 20-year partnership to jointly develop SPARC based technology and systems.
They have had a partnership for 20 years - they aren't announcing a new one.
My guess: Sun drops Ultrasparc III
Sun is already shipping the Ultrasparc IV. Nice guess!
to provide the Throughput computing chips for the low end / web / network stuff,
They have already announced that this is exactly what they are going to do. Again: nice guess!
and takes up the Fujitsu provided SPARC64 chips for the high end and workstation market.
Yesterday's announcement was all about using SPARC64 on the high end. Usually the trick is reading between the lines - not reading the lines themselves.
Sun also announced that they will be using Opterons in their new workstation line - not SPARC64.
Will this spark a new RISC renaissance for Sun and Fujitsu? Or is it a last gasp before Opteron / PowerPC / Itanium crush them?
Itanium has gone white dwarf. The only thing it will be crushing is itself.
Opteron is not going to crush Sun. They have announced that they are shipping multiple Opteron boxes (1-8 way servers and 1-2 way workstations).
This could really revitalize the SPARC system market, especially if Sun's work on Throughput computing proves out.
This doesn't even make sense. The Fujitsu/Sun machines are not the Throughput Computing systems that Sun has been talking about for months. Throughput compututing is Niagara/Rock - the Sun-only CPUs.
No comparison with the other iPod.
:). I'm sure any one of Apple's marketing people knows more about what people will spend than all of /. combined.
You can keep saying that, but that won't make it true. They have the same name, the same functionality, and essentially the same interface. The mini IPod is a little bit smaller than a maxi IPod, but the difference isn't that significant. It's not like it's 1/2 the size.
I don't understand how they think the mini IPod will sell when a "real" Ipod is only $50 more. But, then, I'm just an engineer
The AMD chip, on the other hand, can support x86 relatively easily by including a "morphing layer" (I think that's the name) which maps x86 instructions to the native instructions of the chip
This is only correct if you consider microcode to be the "native instructions" on AMD. Itanic introduced a whole new ISA, which I guess requires some kind of 'morphing to support x86. Opteron uses the existing x86 ISA with a small number of 64-bit extensions. So, x86 is the "native instruction" set for the AMD CPUs, which allows for much better performance of current 32-bit applications.
Is it just me, or does this thing sound exactly like the Cobalt Cube from a few years back?
This is the first first post I've seen that was dumber than a "First post!" first post.
Is it just me, or does this thing sound exactly like the Cobalt Cube from a few years back
It _is_ the Cobalt Qube software. Jesus, I know nobody reads the articles, but is it really too much to ask that you read the title of the message you are responding to?
I certainly wouldn't pay $10/hour if I were planning to be online for many hours. However, if I have a one hour layover in Chicago on a BOS->SFO flight, I might pay $10 to get online for that one hour. If you charge $10 for the first hour and $1 for subsequent hours, then you start making more sense.
I would think that Sun is giving away the x86 Solaris for free because they just want to draw more users from the open source and free-as-in-speech community to look at what it has to offer...
:) These guys are used to seeing money come in in 6-8 figure chunks. I doubt they are staffed to handle lots of tiny little transactions.
They are re-releasing x86 support because they're customers were screaming for it. My guess is that they are giving it away to individual users because it's cheaper than trying to deal with $20 "micropayments."
Absolutely. Unless you've done something stupid like hardcoding some endianness assumption. (of course, that would kill you on any OS.)
Some things aren't supported out of the box, like the ethernet controller (VIA Rhine) built into my motherboard, but there are a surprising number of third-party drivers available online. Wheel mice are also supported by non-Sun drivers.
One real shortcoming is in the area of video drivers. Fortunately, XFree86 works on Solaris x86, addressing that problem rather nicely.
The biggest problem I've run into thus far is a lack of support for Serial ATA. I expect that this is just a result of Solaris x86 being put on hold for a year or so. I can't imagine this wouldn't be available by Solaris 10 (or whatever they end up calling it).
Opteron is cheaper, faster, and requires less power than Itanium. AMD's 64-bit ISA is a hell of a lot cleaner than Itanic's. And AMD appears to be making the switch to 90nm technology faster and better than Intel, which will further help their power/cooling story.