Now, some anonimous flogger asks whether engineering students should be forced to suffer through calculus and physics.
Dozens suffer hard hours studying instead of been drunk.
But we actually need engineers, so, lets make it easy and take calculus away... by the way, physics could go away too.
Ops, now I remember, if we do that instead of engineers they will be technicians.
>Part of the Ethernet spec is to wait 9.6 microseconds after the medium appears to be idle before sending, then resend if it collides. Light moves about 3 kilometres in that time. Making an Ethernet of 2550 km pratically guarantees nothing but collisions. So while this is a hunkin' heap of net, it's not Ethernet.
In the specs of Gigabit Ethernet, there is a lot that has been taken out of Ethernet to avoid overload and get more data transfered, one of the things is the Collision Detection (the one that states how long you have to be idle, the backoff algorithm if there is a collision) in the protocol CSMA/CD (Carrier Sense Multiple Access with Collision Detection), At Gigabit speeds there is no way to think in a common channel (like old days Ethernet where existed coaxial cable), and so CD has no reason to be, also, the channels are now much (much) more reliable than before, so the specs take advantage of that too (and there are minimum quality requirements for the net to work properly).
At Gigabit speeds you do not have traditional old days Ethernet, you have dedicated channels (switches act just like that, and now a switches are almost as cheap as hubs) or multiplexed ones and your Eth card does not need even to know its frames are being multiplexed with others. Data loss at those rates comes from other sources, not from collisions, but from things like EM interference (if you have a copper channel), or package discarding because the client can not handle that much data rate, a broken channel or an overflow of data going into a switch or router.
There is another problem, and is actually the bottleneck of transmitting packets at high rates.
It doesn't really matters (yet, and considering Ethernet technology) if the BW of the fiber is a zillion Petabits/sec. The problem is now at 1Gbps and 10Gbps in Ethernet technology, and is because the processor overloads with the amount of hardware interrupts. The processors that are general purpose have to waste too many clock cycles processing that much interrupts, the processors nowadays are superscalar [ http://en.wikipedia.org/wiki/Superscalar ]and every time the processor have to change the context (to attend an interrupt) has to do lots of things like unloading the registers, saving the context, loading the registers of the new process, and has to drop something out of the pipeline [ http://en.wikipedia.org/wiki/Pipeline_(computing) ] loosing performance.
Ethernet tech has a huge latency [ http://en.wikipedia.org/wiki/Latency_(engineering) ] and a stack that makes processing not so easy (if you look at te code of a linux network device driver it handles pretty much everything including writing the mac address that is only copied when the driver initialize).
That is why there are some relative new things (NAPI in Linux) that try to make lessen the overload, there are new network devices that handle layer 2 and 3 (or at least parts of those, for example, is used to be handled the checksum algorithm) to avoid doing it in the processor. There are some white papers (one from intel, another from NetXen, I'm sorry I don't have the links now) that explain the problem and some approach to a possible solution.
Yes, I know, there is something I have not said, and is that the main switches or routers have to deal with that and have hardware specially designed to do heavy network packet processing, and that is the point, the network cards will have to do that (and are already starting to), neither is an easy job for hardware designers, nor for the market, is easier and cheaper to have a machine that you can change the behaviour only changing the firmware or changing settings from a program (routers have an operating system, and lots of those are a general purpose microprocessor with a linux kernel and a web server to configure it, for example home routers).
I remember from a local air channel news, I was pretty sure was he, but started to look for it after receiving your response. Called a friend that knows more than me about this. The result, I could not find something of substance, but my friend told me two things, according to him I made a mistake, Cohelo just took ideas from others while Bucay is the one I was confused to (and the one that took an entire chapter of Mónica Cavallé), so I regret my mistake ask for sorry to everyone (Cohelo first) and here post some substance about:
http://es.shvoong.com/books/novel/1743054-el-zahir/ The first part of the article
"El Zahir:
No se puede empezar esta sinopsis sin hacer referencia obligada a Jorge Luis Borges, de quien Coelho copia o pide prestado el título de esta novela."
a small translation about the first part:
"The Zahir:
Can't be started a synopsis without referencing Jorge Luis Borges, from who Cohelo copy or borrows the title of this novel"
Some things about Bucay:
http://www.elojodigital.com/sociedad/2005/10/11/739.html http://www.google.com/search?q=bucay+copia+libros+de+espa%C3%B1ola&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:es-AR:official&client=firefox-a
My Mistake not checking data before posting, my karma will fall down the stairs now and will think tree times before posting next time:S
This Autor is very well known here in Argentina for his "self-help books". He writes tons of books,and he doesn't even write them. Why?, well, because he is known to copy from other books and use that in his own work. Some years ago a female spanish autor realized that Coelho was copying from her work (whole chapters) and started legal measures against Paulo.
By the way To pirate is to assault ships in the sea, not copying something.
Slashdot Mirror
Now, some anonimous flogger asks whether engineering students should be forced to suffer through calculus and physics. ... by the way, physics could go away too.
Dozens suffer hard hours studying instead of been drunk.
But we actually need engineers, so, lets make it easy and take calculus away
Ops, now I remember, if we do that instead of engineers they will be technicians.
In this case, GNOME project has some guidelines:
http://library.gnome.org/devel/hig-book/stable/
http://developer.gnome.org/projects/gap/guide/gad/
http://live.gnome.org/Accessibility/GetInvolved
http://developer.gnome.org/projects/gup/
>Part of the Ethernet spec is to wait 9.6 microseconds after the medium appears to be idle before sending, then resend if it collides. Light moves about 3 kilometres in that time. Making an Ethernet of 2550 km pratically guarantees nothing but collisions. So while this is a hunkin' heap of net, it's not Ethernet.
In the specs of Gigabit Ethernet, there is a lot that has been taken out of Ethernet to avoid overload and get more data transfered, one of the things is the Collision Detection (the one that states how long you have to be idle, the backoff algorithm if there is a collision) in the protocol CSMA/CD (Carrier Sense Multiple Access with Collision Detection), At Gigabit speeds there is no way to think in a common channel (like old days Ethernet where existed coaxial cable), and so CD has no reason to be, also, the channels are now much (much) more reliable than before, so the specs take advantage of that too (and there are minimum quality requirements for the net to work properly).
At Gigabit speeds you do not have traditional old days Ethernet, you have dedicated channels (switches act just like that, and now a switches are almost as cheap as hubs) or multiplexed ones and your Eth card does not need even to know its frames are being multiplexed with others. Data loss at those rates comes from other sources, not from collisions, but from things like EM interference (if you have a copper channel), or package discarding because the client can not handle that much data rate, a broken channel or an overflow of data going into a switch or router.
>Interrupt loads can be greatly reduced by switching to a polling-driven architecture. See FreeBSD.
New API (NAPI) takes a mixed approach, read: [ http://en.wikipedia.org/wiki/New_API ], and for more information: [ http://www.linux-foundation.org/en/Net:NAPI ].
There is another problem, and is actually the bottleneck of transmitting packets at high rates.
It doesn't really matters (yet, and considering Ethernet technology) if the BW of the fiber is a zillion Petabits/sec.
The problem is now at 1Gbps and 10Gbps in Ethernet technology, and is because the processor overloads with the amount of hardware interrupts. The processors that are general purpose have to waste too many clock cycles processing that much interrupts, the processors nowadays are superscalar [ http://en.wikipedia.org/wiki/Superscalar ]and every time the processor have to change the context (to attend an interrupt) has to do lots of things like unloading the registers, saving the context, loading the registers of the new process, and has to drop something out of the pipeline [ http://en.wikipedia.org/wiki/Pipeline_(computing) ] loosing performance.
Ethernet tech has a huge latency [ http://en.wikipedia.org/wiki/Latency_(engineering) ] and a stack that makes processing not so easy (if you look at te code of a linux network device driver it handles pretty much everything including writing the mac address that is only copied when the driver initialize).
That is why there are some relative new things (NAPI in Linux) that try to make lessen the overload, there are new network devices that handle layer 2 and 3 (or at least parts of those, for example, is used to be handled the checksum algorithm) to avoid doing it in the processor. There are some white papers (one from intel, another from NetXen, I'm sorry I don't have the links now) that explain the problem and some approach to a possible solution.
Yes, I know, there is something I have not said, and is that the main switches or routers have to deal with that and have hardware specially designed to do heavy network packet processing, and that is the point, the network cards will have to do that (and are already starting to), neither is an easy job for hardware designers, nor for the market, is easier and cheaper to have a machine that you can change the behaviour only changing the firmware or changing settings from a program (routers have an operating system, and lots of those are a general purpose microprocessor with a linux kernel and a web server to configure it, for example home routers).
There is much to say yet in this field.
I remember from a local air channel news, I was pretty sure was he, but started to look for it after receiving your response. Called a friend that knows more than me about this. The result, I could not find something of substance, but my friend told me two things, according to him I made a mistake, Cohelo just took ideas from others while Bucay is the one I was confused to (and the one that took an entire chapter of Mónica Cavallé), so I regret my mistake ask for sorry to everyone (Cohelo first) and here post some substance about:
http://es.shvoong.com/books/novel/1743054-el-zahir/
The first part of the article
"El Zahir: No se puede empezar esta sinopsis sin hacer referencia obligada a Jorge Luis Borges, de quien Coelho copia o pide prestado el título de esta novela." a small translation about the first part: "The Zahir: Can't be started a synopsis without referencing Jorge Luis Borges, from who Cohelo copy or borrows the title of this novel"
Some things about Bucay:
http://www.elojodigital.com/sociedad/2005/10/11/739.html
http://www.google.com/search?q=bucay+copia+libros+de+espa%C3%B1ola&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:es-AR:official&client=firefox-a
My Mistake not checking data before posting, my karma will fall down the stairs now and will think tree times before posting next time:S
This Autor is very well known here in Argentina for his "self-help books". He writes tons of books,and he doesn't even write them. Why?, well, because he is known to copy from other books and use that in his own work. Some years ago a female spanish autor realized that Coelho was copying from her work (whole chapters) and started legal measures against Paulo.
By the way To pirate is to assault ships in the sea, not copying something.