From what I know of the current architectures, AMD's solution to main memory access woes (point-to-point bus) seems more sane as soon as more than a couple of processors are installed in the system. Shared bus (as in Intel's solution) seems to require huge caches to operate efficiently, and as we all know, Pentium 4 really does not like pipeline stalls or branch mispredictions.
Let's take a hypothetical example: quad processor systems utilising dual core processors from Intel and AMD.
AMD: each processor (core) talks directly to its local memory block, and via HT links to adjacent processors' memories. Processors do not have to contest for access to the bus and thus memory access is always low-latency, even when accessing remote memory. If built today, HT links would operate at 1 GHz.
Intel: processors share the same bus with each other and memory controller. Any time a processor needs to access memory, it has to wait until the bus frees to ask the memory controller access to main memory. Pipeline stalls happen here if bus is not free when needed. This is compensated with huge L3 caches. As far as I know, current quad processor systems from Intel have bus speeds of around 533 MHz.
So in a nutshell, Intel competes with AMD on a quite level field when the system has 1-2 processors, but as soon as processor count goes up, bus bandwidth becomes an issue with Intel. It shall be interesting to see how Intel attempts to counter that.
What I am getting at with this? Well, those huge 12 MB L3 caches in Intel's future processors sure aren't cheap. They take up lots of silicon and WILL decrease core yields since they've got lots and lots of points of failure. So manufacturing processes really have to be ramped up to allow that at reasonable cost.
Independent vendors have confirmed that there is currently no defence against this type of attack for DSSS based WLANs
This is the same problem as with LA or VHF radio. Only one device can be transmitting at a time on a single frequency band. This stems from the fact that the receivers have to tune to a certain signal and no two signals are likely to be in the same phase, thus the strongest signal will win. Essentially these devices behave as if they are half-duplex, and well-timed (continuous) collisions will cause the whole segment to come down. This is what happens here. Remember the old coaxial 10base ethernet networks? They were vulnerable to the same thing.
The unfortunate fact here is that there is no cure for this kind of misbehaviour. Old devices likely won't be upgradeable (hence no silver bullet). Multi-band hi-speed WiFi (54Mbit+) is not likely to be affected by this attack, but if they operate in compatibility mode they will be brought down, too. Intelligent access points can lessen the effect of this attack but that leaves the older devices out of the communications.
Essentially this requires quite little work on the part of the attacker since no hi-powered transmitters are needed. That fortunately limits the range of the attack, too. I would like to know if anyone could calculate quick estimates as to the affected area with certain wattage transmitters. Anyone?
Hmm, weird. Mathematics, physics, particle physics, basic and advanced chemistry... I've followed those "threads" quite extensively and I've yet to find a blatant error/nonsense. Of course there are lots of stubs (or just placeholders), but that is just waiting to be rectified. You're right, though, when you say that there's too little existing information in many cases.
As for the politics, I cannot comment as I've only contributed occasionally and anonymously. I certainly hope it's not THAT bad as other people have suggested...
Yep, I know. But there are only 3500 articles in Finnish, which sort of reduces its usefulness to us right now:-)
Ah well, it's a good source of proper english, too, so it can be used to reinforce one's grammar and vocabulary provided that one has access to a good dictionary.
If you don't yet know what that is, head there now to see for yourself. You're doing yourself a disservice by not looking:-)
While you're there make sure to contribute to the topics which are marked red (no explanation yet). That way the great service will become even better and we can all benefit from it.
I know of several schools in my country which instruct their students to go look for information there (in case the student knows english, not our native). I suppose it will only get better since academic institutions are beginning to refer to it.
As always, Microsoft is free to do with its software and userbase as it pleases. I tend myself to sway to the opinion that giving updates away to pirates is not the "right thing" to do. On the one hand we have the internet security issue, on the other hand the rights of MS (they have those, too, you know).
Most of the security problems might be solved by more diligent coding practices and much rewriting of code on the part of Microsoft. Then again it might not help much. It all comes down to how well Microsoft wants to protect its userbase (revenue stream) and how much bad PR it is willing to accept.
Still, I think that much if not most of current Windows codebase is so bloated, the only thing that will help is a complete rewrite (we shall see how Longhorn delivers on this issue).
The moral of the story? Don't pirate MS products, not even to spite them. You will only contribute to their userbase and vendor lock-in that way. And they may even be able to get you to pay for Windows in the end, however that might happen:-)
As a happy Gentoo user, I can testify to the usefulness of the system as a whole. Robbins and his crew really have done it right. It is the least the community can do to give something back to him for his hard work.
We shall see how well the rest of the developers can knit the project together during the following months. It shall be interesting to see who will step forward as the new project leader as well.
Slashdot Mirror
Replying to undo a mis-moderation.
This has been mentioned here already!
From what I know of the current architectures, AMD's solution to main memory access woes (point-to-point bus) seems more sane as soon as more than a couple of processors are installed in the system. Shared bus (as in Intel's solution) seems to require huge caches to operate efficiently, and as we all know, Pentium 4 really does not like pipeline stalls or branch mispredictions.
Let's take a hypothetical example: quad processor systems utilising dual core processors from Intel and AMD.
AMD: each processor (core) talks directly to its local memory block, and via HT links to adjacent processors' memories. Processors do not have to contest for access to the bus and thus memory access is always low-latency, even when accessing remote memory. If built today, HT links would operate at 1 GHz.
Intel: processors share the same bus with each other and memory controller. Any time a processor needs to access memory, it has to wait until the bus frees to ask the memory controller access to main memory. Pipeline stalls happen here if bus is not free when needed. This is compensated with huge L3 caches. As far as I know, current quad processor systems from Intel have bus speeds of around 533 MHz.
So in a nutshell, Intel competes with AMD on a quite level field when the system has 1-2 processors, but as soon as processor count goes up, bus bandwidth becomes an issue with Intel. It shall be interesting to see how Intel attempts to counter that.
What I am getting at with this? Well, those huge 12 MB L3 caches in Intel's future processors sure aren't cheap. They take up lots of silicon and WILL decrease core yields since they've got lots and lots of points of failure. So manufacturing processes really have to be ramped up to allow that at reasonable cost.
This is the same problem as with LA or VHF radio. Only one device can be transmitting at a time on a single frequency band. This stems from the fact that the receivers have to tune to a certain signal and no two signals are likely to be in the same phase, thus the strongest signal will win. Essentially these devices behave as if they are half-duplex, and well-timed (continuous) collisions will cause the whole segment to come down. This is what happens here. Remember the old coaxial 10base ethernet networks? They were vulnerable to the same thing.
The unfortunate fact here is that there is no cure for this kind of misbehaviour. Old devices likely won't be upgradeable (hence no silver bullet). Multi-band hi-speed WiFi (54Mbit+) is not likely to be affected by this attack, but if they operate in compatibility mode they will be brought down, too. Intelligent access points can lessen the effect of this attack but that leaves the older devices out of the communications.
Essentially this requires quite little work on the part of the attacker since no hi-powered transmitters are needed. That fortunately limits the range of the attack, too. I would like to know if anyone could calculate quick estimates as to the affected area with certain wattage transmitters. Anyone?
Hmm, weird. Mathematics, physics, particle physics, basic and advanced chemistry... I've followed those "threads" quite extensively and I've yet to find a blatant error/nonsense. Of course there are lots of stubs (or just placeholders), but that is just waiting to be rectified. You're right, though, when you say that there's too little existing information in many cases.
As for the politics, I cannot comment as I've only contributed occasionally and anonymously. I certainly hope it's not THAT bad as other people have suggested...
Yep, I know. But there are only 3500 articles in Finnish, which sort of reduces its usefulness to us right now :-)
Ah well, it's a good source of proper english, too, so it can be used to reinforce one's grammar and vocabulary provided that one has access to a good dictionary.
If you don't yet know what that is, head there now to see for yourself. You're doing yourself a disservice by not looking :-)
While you're there make sure to contribute to the topics which are marked red (no explanation yet). That way the great service will become even better and we can all benefit from it.
I know of several schools in my country which instruct their students to go look for information there (in case the student knows english, not our native). I suppose it will only get better since academic institutions are beginning to refer to it.
As always, Microsoft is free to do with its software and userbase as it pleases. I tend myself to sway to the opinion that giving updates away to pirates is not the "right thing" to do. On the one hand we have the internet security issue, on the other hand the rights of MS (they have those, too, you know).
:-)
Most of the security problems might be solved by more diligent coding practices and much rewriting of code on the part of Microsoft. Then again it might not help much. It all comes down to how well Microsoft wants to protect its userbase (revenue stream) and how much bad PR it is willing to accept.
Still, I think that much if not most of current Windows codebase is so bloated, the only thing that will help is a complete rewrite (we shall see how Longhorn delivers on this issue).
The moral of the story? Don't pirate MS products, not even to spite them. You will only contribute to their userbase and vendor lock-in that way. And they may even be able to get you to pay for Windows in the end, however that might happen
As a happy Gentoo user, I can testify to the usefulness of the system as a whole. Robbins and his crew really have done it right. It is the least the community can do to give something back to him for his hard work.
We shall see how well the rest of the developers can knit the project together during the following months. It shall be interesting to see who will step forward as the new project leader as well.