I think it is possible, though I never used it. It is a bit hidden down in the network settings dialogs, however.
I guess it works like this: (I don't have an english NT, so bear with my translation, please.) - network properties - TCP/IP properties - options on "IP address" pane - check "activate security" - list the allowed ports.
I guess you can do it only interface-based, not source-address-based, though.
You measure system speed only in terms of 3D games performance and you do faulty reasoning. Just because RDRAM has a lousy performance it doesn't prove the memory bandwidth is not a bottleneck.
There are lots of applications where CPU bus and memory bandwith are the bottlenecks. Why do you think some high-end servers provide 4-way interleave access to their SDRAM main memory?
They don't even bother to keep their Linux variant up to date, so I don't think they'll preinstall NetBSD.
After all, these maschines are positioned for the network appliance market, the customers aren't even supposed to know which OS is running on these boxes.
Too bad the SAK does nothing else but killing off every process spawned under that virtual console, so it is not a good idea if you use shell escapes or job control.
Noone does the actual rendering on Onyxes. The actual rendering is trivially parallelizable on distributed memory systems and that clusters in fact are. Onyxs performance is about heavy-duty *real time* visualization. If you get simulation data from a supercomputer and want to view it interactively in a 4-side cave, you need a fast shared memory parrallelism and most important (and what's the real expensive part of an Onyx) four high-end graphics pipelines.
The "Open Source" model is probably the supreme way of getting feedback from the average technical user. This does not mean that it is at all good at getting feedback from the average nontechnical user.
Yep. The author already said that. The question is how to adapt the Open Source model to getting feedback from the average nontechnical user.
This is the mentality that allows many technical people to grin vaguely when Sun talks about replacing our desktop powerhouses with brainless terminals that place the power of public speech back in the hands of a rich few.
Nonsense. Sun's talks about "network computers" (read: X-Terminals) have nothing to do with dictatorship over the users. They are about manageability in large corporate networks. I, as a network administrator, want the users I support to be able to go to *any* station in the building, log in and get all their apps configured with their personal settings so that they can just start. I want to be able to just plug a new station into the net and it immediately get configured just like the others around. Try to do this with a standard Windows or Linux setup. (I admit I haven't actually implemented this with X-Terminals or the Sun stuff, but it sounds more promising than anything else)
If you are afraid of losing your "desktop powerhouse" think about how much you can do with a common PC setup found in any larger company with the network unplugged. You won't be able to even log in.
Sun won't be taking away your home desktop PC. I you are believing this, you are believing marketing rubbish. It is about the corporate desktop and about how to make the admin's life easier while the user doesn't care or even doesn't notice.
Because you asked: SGI's current processors are clocked from 225 to 300 MHz, while Sun's UltraSparc II(i) is clocked from 248 to 480MHz.
But you recognized it: clock ratings are little more than marketing hype. As you can see from the AMD-vs-Intel battle and Cyrix's PR-ratings you can't even compare MHz-ratings of x86 processors, let alone those of different architectures.
Furthermore, in the enterprise-class server market it isn't all about CPU speed, it is I/O throughput which is way more interesting.
I thought they'd use a backend more like that of MS SQL Server. Since they position Exchange for enterprise use, it would be a bit more appropriate than Jet DB, which doesn't scale well beyond a small workgroup.
Simple answer: Because the users want them. You could have mountpoint-like behaviour with DOS using the JOIN command, you could have mountpoint-like behaviour with NT 4 using the DFS extension.
Nobody used it. And that's why we still have drive letters.
Siemens seems to be phasing out their MIPS-based RMxxx series of workstations and servers, and is switching to a UltraSPARC-based server line developed together with Fujitsu which will be running Solaris. (announced end of last year, Fujitsu and Siemens work together quite a bit, e.g. in the PC market as well) I guess SINIX will die along with the MIPS-based RM machines.
Is the x86 variant of SINIX really running on PCs? I always thought is was running exclusively on the MX300, a 386/486-based machine form Siemens which is not PC-compatible.
BTW: Thanks for the pointer to the online manual. I happen to have a RM400 running SINIX V5.42 at home, but I didn't have a manual. (So at least I am glad about a SINIX version of Netscape Navigator, too:-)
Slashdot Mirror
Aren't you mixing RAID and logging/journaling file systems?
I guess every interleaving memory access requires the memory in pairs (or even quadruples for 4-way interleaving)
Probably the same one who made the Microchannel architecture that tremendous success.
SCNR
Wasn't NT 4.0 C2 certified with a network connection just a feq months ago?
I think it is possible, though I never used it.
It is a bit hidden down in the network settings dialogs, however.
I guess it works like this:
(I don't have an english NT, so bear with my translation, please.)
- network properties
- TCP/IP properties
- options on "IP address" pane
- check "activate security"
- list the allowed ports.
I guess you can do it only interface-based,
not source-address-based, though.
HTH
From what I heard about Photoshop for Unix from our chief graphics designer,
it sucked anyway compared to both Mac and Windows versions of that time.
AFAIK it is proven that any higher-times DES give no significant security bonus over TripleDES regardless of the greater keylength.
If you are sure it is a real bug, you could post to the linux-kernel mailing list.
They are NOT.
Real 1GHz PIIIs are not yet available.
There are lots of applications where CPU bus and memory bandwith are the bottlenecks. Why do you think some high-end servers provide 4-way interleave access to their SDRAM main memory?
... if AMD finally made SMP-capable Athlons.
I guess there are Alpha-based EV6 SMP boards available, aren't they?
They don't even bother to keep their Linux variant up to date, so I don't think they'll preinstall NetBSD.
After all, these maschines are positioned for the network appliance market, the customers aren't even supposed to know which OS is running on these boxes.
I haven't seen any PCI network card requiring manual setup of IRQs or I/O adresses.
Yeah, right. And everyone running a Unix server
is told not to run an X11 server on it.
Where's the difference?
You'd have to read and understand it.
With a networked scheme, the server would do all this for you.
Too bad the SAK does nothing else but killing off every process spawned under that virtual console,
so it is not a good idea if you use shell escapes or job control.
They are doing QoS for Napster already.
Napster gets assigned quality 0: drop entirely.
Noone does the actual rendering on Onyxes.
The actual rendering is trivially parallelizable on distributed memory systems and that clusters in fact are. Onyxs performance is about heavy-duty *real time* visualization.
If you get simulation data from a supercomputer and want to view it interactively in a 4-side cave, you need a fast shared memory parrallelism and most important (and what's the real expensive part of an Onyx) four high-end graphics pipelines.
Yep. The author already said that.
The question is how to adapt the Open Source model to getting feedback from the average nontechnical user.
Nonsense.
Sun's talks about "network computers" (read: X-Terminals) have nothing to do with dictatorship over the users. They are about manageability in large corporate networks.
I, as a network administrator, want the users I support to be able to go to *any* station in the building, log in and get all their apps configured with their personal settings so that they can just start. I want to be able to just plug a new station into the net and it immediately get configured just like the others around. Try to do this with a standard Windows or Linux setup.
(I admit I haven't actually implemented this with X-Terminals or the Sun stuff, but it sounds more promising than anything else)
If you are afraid of losing your "desktop powerhouse" think about how much you can do with a common PC setup found in any larger company with the network unplugged. You won't be able to even log in.
Sun won't be taking away your home desktop PC. I you are believing this, you are believing marketing rubbish. It is about the corporate desktop and about how to make the admin's life easier while the user doesn't care or even doesn't notice.
Because you asked:
SGI's current processors are clocked from 225 to 300 MHz, while Sun's UltraSparc II(i) is clocked from 248 to 480MHz.
But you recognized it:
clock ratings are little more than marketing hype.
As you can see from the AMD-vs-Intel battle and Cyrix's PR-ratings you can't even compare MHz-ratings of x86 processors, let alone those of different architectures.
Furthermore, in the enterprise-class server market it isn't all about CPU speed, it is I/O throughput which is way more interesting.
Do the really use the Jet DB?
Oh, dear.
I thought they'd use a backend more like that of MS SQL Server. Since they position Exchange for enterprise use, it would be a bit more appropriate than Jet DB, which doesn't scale well beyond a small workgroup.
Simple answer: Because the users want them.
You could have mountpoint-like behaviour with DOS using the JOIN command,
you could have mountpoint-like behaviour with NT 4 using the DFS extension.
Nobody used it.
And that's why we still have drive letters.
The MKS toolkit has a working ln as well.
I wonder if one is included with the NT Server Ressource Kit, too.
Siemens seems to be phasing out their MIPS-based RMxxx series of workstations and servers, and is switching to a UltraSPARC-based server line developed together with Fujitsu which will be running Solaris.
:-)
(announced end of last year, Fujitsu and Siemens work together quite a bit, e.g. in the PC market as well)
I guess SINIX will die along with the MIPS-based RM machines.
Is the x86 variant of SINIX really running on PCs?
I always thought is was running exclusively on the MX300, a 386/486-based machine form Siemens which is not PC-compatible.
BTW: Thanks for the pointer to the online manual. I happen to have a RM400 running SINIX V5.42 at home, but I didn't have a manual.
(So at least I am glad about a SINIX version of Netscape Navigator, too